Methods of detecting over-immunosuppression and under- immunosuppression in renal transplant recipients

Abstract

Compositions, methods, kits, systems, and software are provided for detecting over-immunosuppression or under-immunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy. In particular, the methods utilize a urine-based metabolite signature that differentiates over-immunosuppressed or under-immunosuppressed renal transplant recipients from those with a stable graft with high accuracy. In some embodiments, nuclear magnetic resonance spectroscopy and / or liquid chromatography-mass spectrometry techniques are used to detect metabolite features that differentiate patients who are over-immunosuppressed or under-immunosuppressed from those with stable grafts. The subject methods can be used for monitoring a renal transplant recipient to avoid over-immunosuppression, which can lead to infections that cause graft dysfunction or loss or even death, as well as monitoring a renal transplant recipient to avoid under-immunosuppression, which can lead to graft dysfunction, subclinical graft rejection, or rejection.

Description

METHODS OF DETECTING OVER-IMMUNOSUPPRESSION AND UNDER¬IMMUNOSUPPRESSION IN RENAL TRANSPLANT RECIPIENTSCROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims benefit of U.S. Provisional Patent Application No. 63 / 653,535, filed May 30, 2024, and U.S. Provisional Patent Application No. 63 / 591 ,949, filed October 20, 2023, which applications are incorporated herein by reference in their entireties.BACKGROUND OF THE INVENTION

[0002] For the over 700,000 CKD patients in the U.S. who have progressed to end-stage renal disease (ESRD) (USRDS, 2021), kidney transplant (KT) is a life-saving option, with dramatically improved quality of life compared to dialysis with over 87% of patients surviving beyond 5 years (USRDS, 2021 ). As of 2019, there were 239,413 patients in the United States with a functioning KT, and there were over 25,000 KTs performed in 2022, setting an all-time record since transplantation began (United Network for Organ Sharing (UNOS), 2023; USRDS, 2021 ; J. H. Wang & Hart, 2021 ).

[0003] The average billed amount for a KT in the United States in 2020 was $442,500 (T. Scott Bentley and Nick Ortner, 2020), making the clinical and economic effort to prevent graft loss paramount for both renal transplant recipients (RTRs) and the healthcare system overall. As part of this effort, RTRs must receive lifelong doses of immunosuppressants to prevent graft rejection. Achieving immunosuppressant dose stability, or what we term “immunostability,” is a delicate balance between the extremes and remains a clinical challenge (Wojciechowski & Wiseman, 2021 ). Too little treatment leads to under-immunosuppression, risking graft rejection, while too much leads to over-immunosuppression, resulting in opportunistic infection and malignancy.

[0004] Modern immunosuppressive regimens have improved rejection outcomes (Durrbach et aL, 2010), with one-year rejection rates decreasing to <10% after the emergence of a combined tacrolimus / cyclosporine / mycophenolic acid regimen in the late 1990s (Nelson et aL, 2022) and an all-time high adjusted 5-year kidney survival reported in 2020 (USRDS, 2021 ). However, over the same period, these same immunosuppressant agents have been identified as significant risk factors for infection (Min et aL, 2010) and malignancy (Rama & Grinyo, 2010).

[0005] Post-transplant infections stand as the second-leading cause of death with functioning graft (DWFG) in RTRs within the first year (Hariharan et aL, 2021 ), and 78% of RTRs experience overimmunosuppression within 5 years of KT (Jackson et aL, 2021). One common result of overimmunosuppression is polyomavirus-associated nephropathy (PVAN), typically caused by opportunistic activation of BK virus (BKV), a ubiquitous polyomavirus residing in the renal tubularinterstitial epithelium (Kuypers, 2012). PVAN occurs in 5-10% of RTRs and can lead to graft dysfunction or loss (Wojciechowski & Wiseman, 2021 ). BKV is readily detected by PCR based methods, however because it is present in 80-90% of normal adults, the clinical utility of the assay to predict BKVIN remains limited (Egli, 2009). Nonetheless if diagnosed early, BKVIN may be resolved by reducing immunosuppressant levels. However, this can create a vicious cycle, wherein the reduction of therapy to manage the infection, triggers the production of donor specific antibodies (DSAs), and then a subsequent increase in immunosuppressants to prevent rejection leads to a viral flare up. It is a challenge for clinicians to find the delicate balance between under- and overimmunosuppression for RTRs.

[0006] At present, the primary method to assess KT function is by comparing serial serum creatine levels. However, as creatine levels are subject to change due to a variety of other circumstances, it is not a sufficient metric to monitor graft function. Other methods for monitoring kidney function include 24-hour urine collections, proteinuria, and proxies for measuring glomerular filtration rate (GFR) such as iohexol, iothalamate, or inulin clearance (Josephson, 2011 ). However, these tests are less practical, more expensive, and less readily available. Therapeutic drug monitoring (TDM) is also often used to ensure immunosuppressant levels are within an expected range in an RTRs patient’s blood. However, the therapeutic ranges are based on a mixture of empirical observations from a limited number of patients and have come under revision (Traitanon, 2014). Measurement of DSAs and / or and donor-derived cell-free DNA (dd-cfDNA) in the blood is also used to evaluate graft function and risk of rejection (Rogulska et al., 2022)(Oellerich et al., 2021 ). Unfortunately, these assays have incredibly low predictive power (50-60%) to detect under-immunosuppression and no ability to detect over-immunosuppression (Lamarche et al., 2016; Oellerich et al., 2021 ).

[0007] Biopsy remains the only method to fully assess graft function (Cosio, 2005). However, biopsies add significant cost to the healthcare system and utilization greatly varies from center to center with only 17% of transplant centers perform protocol biopsy (Mehta et al., 2017), and no standard criteria to trigger a “for cause” biopsy. Further, biopsy is an invasive procedure, posing a significant risk to already immunocompromised RTRs. Further, biopsy results may not clearly determine the underlying issue (Rogulska et aL, 2022). For example, a biopsy finding of overt tubular necrosis may be indicative of either BKV infection (signaling under-immunosuppression) or antibody- mediated rejection (signaling under-immunosuppression) under the current Banff criteria (Jeong, 2020). Further, intra- and inter-scorer variability is high with allograft biopsy scoring, resulting in poor reproducibility for biopsy-based scoring of graft dysfunction post-KT (Huang et aL, 2023) Thus, even biopsy may leave clinicians without the insight to proceed with confidence to adjust medicationdosage. There is a critical need for non-invasive biomarkers to monitor under and overimmunosuppression in RTRs.SUMMARY OF THE INVENTION

[0008] Compositions, methods, kits, systems, and software are provided for detecting overimmunosuppression or under-immunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy. In particular, the methods utilize a urine-based metabolite signature that differentiates over-immunosuppressed or under-immunosuppressed renal transplant recipients from those with a stable graft with high accuracy. In some embodiments, nuclear magnetic resonance spectroscopy and / or liquid chromatography-mass spectrometry techniques are used to detect metabolite features that differentiate patients who are over-immunosuppressed or under-immunosuppressed from those with stable grafts. The subject methods can be used for monitoring a renal transplant recipient to avoid over-immunosuppression, which can lead to infections that cause graft dysfunction or loss or even death, as well as monitoring a renal transplant recipient to avoid under-immunosuppression, which can lead to graft dysfunction, subclinical graft rejection, or rejection.

[0009] In one aspect, a method of detecting over-immunosuppression in a renal transplant recipient undergoing immunosuppressive therapy is provided, the method comprising: (a) obtaining a urine sample from the renal transplant recipient; (b) measuring levels of one or more metabolites in the urine; and (c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed.

[0010] In certain embodiments, the method further comprises altering the immunosuppressive therapy (e.g., by reducing dosage of an immunosuppressive agent or changing the type of immunosuppressive agent administered to the renal transplant recipient) if the renal transplant recipient is determined to be over-immunosuppressed.

[0011] In certain embodiments, the method further comprises performing diagnostic screening to determine whether the renal transplant recipient has an infection, a malignancy, or polyomavirus- associated nephropathy if the renal transplant recipient is determined to be overimmunosuppressed; and treating the renal transplant recipient for the infection, the malignancy, or the polyomavirus-associated nephropathy if the renal transplant recipient is diagnosed as having the infection, the malignancy, or the polyomavirus-associated nephropathy. In some embodiments, the infection is a BK virus infection.

[0012] In certain embodiments, the levels of one or more metabolites in the urine are measured using nuclear magnetic resonance (NMR) spectroscopy. In some embodiments, the NMRspectroscopy is multi-dimensional NMR spectroscopy. In some embodiments, the NMR spectroscopy comprises one-dimensional (1 D)1H NMR spectroscopy, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectroscopy, or a combination thereof. In some embodiments, the NMR spectroscopy is performed with non-uniformed sampling (NUS).

[0013] In certain embodiments, one or more differential NMR metabolite features are used to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model, wherein the one or more differential NMR metabolite features are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.94 13C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.

[0014] In certain embodiments, the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1 103 having a chemical shift at 3.961H ppm and a chemical shift at 72.85 13C ppm.

[0015] In certain embodiments, the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.06 ’H ppm and a chemical shift at 85.9413C ppm.

[0016] In certain embodiments, the method further comprises using one or more normalization NMR metabolite features in combination with the differential NMR metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

[0017] In certain embodiments, the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.55 ’H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shiftat 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71.9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm. In some embodiments, the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

[0018] In certain embodiments, the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.

[0019] In certain embodiments, the chemical shifts are calibrated against an internal deuterated sodium 2,2-dimethyl-2-silapentane-5-sulfonate standard.

[0020] In certain embodiments, the levels of one or more metabolites in the urine are measured using mass spectrometry. In some embodiments, the levels of one or more metabolites in the urine are measured using liquid chromatography-mass spectrometry (LC-MS).

[0021] In certain embodiments, ionization is performed using heated electrospray ionization (HESI).

[0022] In certain embodiments, the liquid chromatography is performed using a linear gradient of5% to 95% acetonitrile.

[0023] In certain embodiments, one or more differential LC-MS metabolite features of caffeine, benzoyl formic acid, and cytidine are used to classify the renal transplant recipient as overimmunosuppressed or stable using the machine learning model. In some embodiments, the differential LC-MS metabolite features for caffeine comprise a feature with a mass-to-charge ratio (m / z) of 195.00 for a caffeine ([M+H]+ ion and a feature with a m / z of 137.90 and a feature with a m / z of 110.00 from selected reaction monitoring. In some embodiments, the differential LC-MS metabolite features for benzoyl formic acid comprise a feature with a m / z of 149.09 for a benzoyl formic acid [M-H]- ion and a feature with a m / z of 104.91 and a feature with a m / z of 77.07 from selected reaction monitoring. In some embodiments, the differential LC-MS metabolite features forcytidine comprise a feature with a m / z of 244.09 for a cytidine [M+H]+ ion and a feature with a m / z of 111 .90 from selected reaction monitoring.

[0024] In certain embodiments, the method further comprises using one or more normalization LC- MS metabolite features in combination with the differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

[0025] In certain embodiments, the normalization LC-MS metabolite features comprise lysine, hippuric acid, mannitol, trimethylamine N-oxide, and glutamine LC-MS features.

[0026] In certain embodiments, the one or more differential NMR metabolite features are used in combination with one or more differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model, wherein the one or more differential NMR metabolite features are selected from X109 having a chemical shift at 3.01 ’H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm; and wherein the one or more differential LC-MS metabolite features are selected from a feature with a mass-to-charge ratio (m / z) of 195.00 for a caffeine ([M+H]+ ion and a feature with a m / z of 137.90 and a feature with a m / z of 1 10.00 from selected reaction monitoring, a feature with a m / z of 149.09 for a benzoyl formic acid [M-H]- ion and a feature with a m / z of 104.91 and a feature with a m / z of 77.07 from selected reaction monitoring, a feature with a m / z of 244.09 for a cytidine [M+H]+ ion and a feature with a m / z of 111 .90 from selected reaction monitoring.

[0027] In certain embodiments, the differential NMR metabolite features, used in combination with one or more differential LC-MS metabolite features, comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.

[0028] In certain embodiments, the differential NMR metabolite features, used in combination with one or more differential LC-MS metabolite features, comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.

[0029] In certain embodiments, the method further comprises using one or more normalization NMR metabolite features and normalization LC-MS metabolite features in combination with the differential NMR metabolite features and differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

[0030] In certain embodiments, the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71.9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

[0031] In certain embodiments, the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.72 13C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.83 ’H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.78 1H ppm and a chemical shift at 57.0813C ppm.

[0032] In certain embodiments, the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.

[0033] In certain embodiments, the normalization LC-MS metabolite features comprise lysine, hippuric acid, mannitol, trimethylamine N-oxide, and glutamine LC-MS features.

[0034] In certain embodiments, the method further comprises isolating metabolites from the urine sample prior to said measuring.

[0035] In certain embodiments, the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA), or a random forest (RF) machine learning algorithm.

[0036] In certain embodiments, the method further comprises repeating steps (a) - (c) periodically to determine whether the renal transplant recipient is becoming over-immunosuppressed over time. In some embodiments levels of metabolite biomarkers are measured in urine samples at set intervals. In some embodiments, the levels of metabolite biomarkers are measured at least twice a month, at least once a month, at least every 2 months, at least every 3 months, at least every 4 months, at least every 5 months, at least every 6 months, or at least once a year and analyzed, as described herein, to determine if the renal transplant recipient is becoming over-immunosuppressed. In some embodiments, levels of metabolite biomarkers are measured in urine samples if the subject shows symptoms of an infection or transplant rejection.

[0037] In another aspect, a biomarker selected from lysine, caffeine, benzoyl formic acid, and cytidine for use in a method of diagnosing over-immunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy is provided.

[0038] In another aspect, a kit is provided, the kit comprising: a container for collecting a urine sample from a renal transplant recipient, and instructions to halt, alter, or monitor treatment of the renal transplant recipient based on analyzing the levels of one or more metabolites in the urine sample using a machine learning model to determine whether the renal transplant recipient is overimmunosuppressed according to a method described herein.

[0039] In another aspect, a computer implemented method for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed is provided, the computer performing steps comprising: a) receiving NMR and / or LC-MS spectra of one or more metabolites from a urine sample obtained from the renal transplant recipient; b) measuring levels of the one or more metabolites using the NMR and / or LC-MS spectra; c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed; and d) displaying information regarding whether the renal transplant recipient is over-immunosuppressed.

[0040] In certain embodiments, the computer implemented further comprises storing the information regarding whether the renal transplant recipient is over-immunosuppressed in a database.

[0041] In certain embodiments, the NMR spectra are multi-dimensional NMR spectra.

[0042] In certain embodiments, the NMR spectra comprise one-dimensional (1 D)1H NMR spectra,2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectra, or a combination thereof.

[0043] In certain embodiments, the NMR spectra are obtained using non-uniformed sampling (NUS).

[0044] In certain embodiments, one or more differential NMR metabolite features in the NMR spectra are used to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model, wherein the one or more differential NMR metabolite features are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.28 ’H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.

[0045] In certain embodiments, the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.85 13C ppm.

[0046] In certain embodiments, the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.

[0047] In certain embodiments, the computer implemented method further comprises using one or more normalization NMR metabolite features in combination with the differential NMR metabolite features in the NMR spectra to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

[0048] In certain embodiments, one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.55 ’H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm,X63 having a chemical shift at 3.831H ppm and a chemical shift at 71.9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

[0049] In certain embodiments, the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.72 13C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71.9313C ppm, X83 having a chemical shift at 3.28 ’H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.78 ’H ppm and a chemical shift at 57.0813C ppm.

[0050] In certain embodiments, the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.

[0051] In certain embodiments, the chemical shifts are calibrated against an internal deuterated sodium 2,2-dimethyl-2-silapentane-5-sulfonate standard.

[0052] In certain embodiments, one or more differential LC-MS metabolite features of caffeine, benzoyl formic acid, and cytidine in the LC-MS spectra are used to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

[0053] In certain embodiments, the differential LC-MS metabolite features for caffeine comprise a feature with a mass-to-charge ratio (m / z) of 195.00 for a caffeine ([M+H]+ ion and a feature with a m / z of 137.90 and a feature with a m / z of 110.00 from selected reaction monitoring.

[0054] In certain embodiments, the differential LC-MS metabolite features for benzoyl formic acid comprise a feature with a m / z of 149.09 for a benzoyl formic acid [M-H]- ion and a feature with a m / z of 104.91 and a feature with a m / z of 77.07 from selected reaction monitoring.

[0055] In certain embodiments, the differential LC-MS metabolite features for cytidine comprise a feature with a m / z of 244.09 for a cytidine [M+H]+ ion and a feature with a m / z of 111 .90 from selected reaction monitoring.

[0056] In certain embodiments, the computer implemented method further comprises using one or more normalization LC-MS metabolite features in combination with the differential LC-MS metabolite features in the LC-MS spectra to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

[0057] In certain embodiments, the normalization LC-MS metabolite features comprise lysine, hippuric acid, mannitol, trimethylamine N-oxide, and glutamine LC-MS features.

[0058] In certain embodiments, the one or more differential NMR metabolite features are used in combination with one or more differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model, wherein the one or more differential NMR metabolite features are selected from X109 having a chemical shift at 3.01 ’H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm; and wherein the one or more differential LC-MS metabolite features are selected from a feature with a mass-to-charge ratio (m / z) of 195.00 for a caffeine ([M+H]+ ion and a feature with a m / z of 137.90 and a feature with a m / z of 110.00 from selected reaction monitoring, a feature with a m / z of 149.09 for a benzoyl formic acid [M-H]- ion and a feature with a m / z of 104.91 and a feature with a m / z of 77.07 from selected reaction monitoring, a feature with a m / z of 244.09 for a cytidine [M+H]+ ion and a feature with a m / z of 111 .90 from selected reaction monitoring.

[0059] In certain embodiments, the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.85 13C ppm.

[0060] In certain embodiments, the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.

[0061] In certain embodiments, the computer implemented method further comprises using one or more normalization NMR metabolite features and normalization LC-MS metabolite features in combination with the differential NMR metabolite features and differential LC-MS metabolite featuresto classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

[0062] In certain embodiments, the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71.9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

[0063] In certain embodiments, the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.72 13C ppm, X59 having a chemical shift at 5.24 ’H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.78 1H ppm and a chemical shift at 57.0813C ppm.

[0064] In certain embodiments, the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.

[0065] In certain embodiments, the normalization LC-MS metabolite features comprise lysine, hippuric acid, mannitol, trimethylamine N-oxide, and glutamine LC-MS features.

[0066] In certain embodiments, the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA) or a random forest (RF) machine learning algorithm.

[0067] In another aspect, a system for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed using a computer implemented method, described herein, is provided, the system comprising: a) a storage component for storing data, wherein the storage component has instructions for evaluating whether a renaltransplant recipient undergoing treatment with immunosuppressive therapy is overimmunosuppressed based on analysis of the NMR and / or LC-MS spectra of the one or more metabolites stored therein; b) a computer processor programmed to analyze the NMR and / or LC- MS spectra using one or more algorithms, wherein the computer processor is coupled to the storage component and configured to execute the instructions stored in the storage component in order to receive the inputted NMR and / or LC-MS spectra, and analyze the NMR and / or LC-MS spectra according to a computer implemented method described herein; and c) a display component for displaying information regarding whether the renal transplant recipient is over-immunosuppressed.

[0068] In another aspect, a non-transitory computer-readable medium comprising program instructions that, when executed by a processor in a computer, causes the processor to perform the computer implemented method, described herein, is provided.

[0069] In another aspect, a kit comprising the non-transitory computer-readable medium, described herein, and instructions for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed is provided.

[0070] In another aspect, a method of detecting over-immunosuppression or underimmunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy is provided, the method comprising: (a) obtaining a urine sample from the renal transplant recipient; (b) measuring levels of one or more metabolites in the urine; and (c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed or under-immunosuppressed.

[0071] In certain embodiments, the one or more metabolites are selected from lysine, caffeine, benzoyl formic acid, cytidine, glucuronic acid, glucaric acid, lactose, glucose-6-phosphate, 1 - methylhistamine, histidine, trigonelline, trimethylamine N-oxide (TMAO), hippuric acid, choline, phosphoethanolamine, phosphocholine, acetylglycine, phenyllactic acid, mycophenolic acid glucuronide (MPAG), creatine, and hydroxyphenylacetic acid.

[0072] In certain embodiments, the method further comprises altering the immunosuppressive therapy administered to the renal transplant recipient if the renal transplant recipient is determined to be over-immunosuppressed or under-immunosuppressed, wherein immunosuppression is decreased if the renal transplant recipient is determined to be over-immunosuppressed and increased if the renal transplant recipient is determined to be under-immunosuppressed.

[0073] In certain embodiments, the method further comprises performing diagnostic screening to determine whether the renal transplant recipient has an infection, a malignancy, or polyomavirus- associated nephropathy if the renal transplant recipient is determined to be overimmunosuppressed; and treating the renal transplant recipient for the infection, the malignancy, orthe polyomavirus-associated nephropathy if the renal transplant recipient is diagnosed as having the infection, the malignancy, or the polyomavirus-associated nephropathy.

[0074] In certain embodiments, the method further comprises performing diagnostic screening to determine whether the renal transplant recipient has graft dysfunction, subclinical graft rejection, or graft rejection if the renal transplant recipient is determined to be under-immunosuppressed; and treating the renal transplant recipient for the graft dysfunction, subclinical graft rejection, or graft rejection.

[0075] In certain embodiments, the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA) or a random forest (RF) machine learning algorithm.

[0076] In certain embodiments, the levels of one or more metabolites in the urine are measured using mass spectrometry.

[0077] In certain embodiments, the levels of one or more metabolites in the urine are measured using nuclear magnetic resonance (NMR) spectroscopy. In some embodiments, the NMR spectroscopy is multi-dimensional NMR spectroscopy. In some embodiments, the NMR spectroscopy comprises one-dimensional (1 D)1H NMR spectroscopy, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectroscopy, or a combination thereof. In some embodiments, the NMR spectroscopy is performed with non-uniformed sampling (NUS).

[0078] In certain embodiments, one or more differential NMR metabolite features are used to classify the renal transplant recipient as over-immunosuppressed, under-immunosuppressed, or stable using the machine learning model, wherein the one or more differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1 103 having a chemical shift at 3.961H ppm and a chemical shift at 72.85 13C ppm; and wherein the one or more differential NMR metabolite features used to classify the renal transplant recipient as under-immunosuppressed or stable are selected from X65 having a chemical shift at 4.06 ’H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X1223 having a chemical shift at 7.941H ppm and a chemical shift at 140.3013C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 havinga chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.

[0079] In certain embodiments, the differential NMR metabolite features, used to classify the renal transplant recipient as over-immunosuppressed or stable, comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.

[0080] In certain embodiments, the differential NMR metabolite features, used to classify the renal transplant recipient as over-immunosuppressed or stable, comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.

[0081] In certain embodiments, the differential NMR metabolite features, used to classify the renal transplant recipient as under-immunosuppressed or stable, comprise or consist of X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.67 ’H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51 .0413C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.55 13C ppm.

[0082] In certain embodiments, the method further comprises using one or more normalization NMR metabolite features in combination with the differential NMR metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

[0083] In certain embodiments, the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71.9313C ppm, X83 having achemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

[0084] In certain embodiments, the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.72 13C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.78 ’H ppm and a chemical shift at 57.0813C ppm.

[0085] In certain embodiments, the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.

[0086] In certain embodiments, the chemical shifts are calibrated against an internal deuterated sodium 2,2-dimethyl-2-silapentane-5-sulfonate standard.

[0087] In certain embodiments, the method further comprises repeating steps (a) - (c) periodically to determine whether the renal transplant recipient has become over-immunosuppressed. In some embodiments, steps (a) - (c) are repeated at least twice a month, at least once a month, at least every 2 months, at least every 3 months, at least every 4 months, at least every 5 months, at least every 6 months, or at least once a year.

[0088] In another aspect, a biomarker selected from lysine, caffeine, benzoyl formic acid, cytidine, glucuronic acid, glucaric acid, lactose, glucose-6-phosphate, 1 -methylhistamine, histidine, trigonelline, trimethylamine N-oxide (TMAO), hippuric acid, choline, phosphoethanolamine, phosphocholine, acetylglycine, phenyllactic acid, mycophenolic acid glucuronide (MPAG), creatine, and hydroxyphenylacetic acid for use in a method of diagnosing over-immunosuppression or underimmunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy is provided.

[0089] In another aspect, a kit is provided, the kit comprising: a container for collecting a urine sample from a renal transplant recipient, and instructions to halt, alter, or monitor treatment of the renal transplant recipient based on analyzing the levels of one or more metabolites in the urine sample using a machine learning model to determine whether the renal transplant recipient is overimmunosuppressed according to the method described herein.

[0090] In another aspect, a computer implemented method for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed, underimmunosuppressed, or stable is provided, the computer performing steps comprising:

[0091] A computer implemented method for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed, underimmunosuppressed, or stable, the computer performing steps comprising: (a) receiving nuclear magnetic resonance (NMR) and / or mass spectrometry spectra of one or more metabolites from a urine sample obtained from the renal transplant recipient; (b) measuring levels of the one or more metabolites using the NMR spectra and / or mass spectrometry spectra; (c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed, under-immunosuppressed, or stable; and (d) displaying information regarding whether the renal transplant recipient is over-immunosuppressed, under-immunosuppressed, or stable.

[0092] In certain embodiments, the computer implemented method further comprises storing the information regarding whether the renal transplant recipient is over-immunosuppressed, underimmunosuppressed, or stable in a database.

[0093] In certain embodiments, one or more differential NMR metabolite features are used to classify the renal transplant recipient as over-immunosuppressed, under-immunosuppressed, or stable using the machine learning model, wherein the one or more differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1 103 having a chemical shift at 3.961H ppm and a chemical shift at 72.85 13C ppm; and wherein the one or more differential NMR metabolite features used to classify the renal transplant recipient as under-immunosuppressed or stable are selected from X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X1223 having a chemical shift at 7.941H ppm and a chemical shift at 140.3013C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.

[0094] In certain embodiments, the differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.

[0095] In certain embodiments, the differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.

[0096] In certain embodiments, the differential NMR metabolite features used to classify the renal transplant recipient as under-immunosuppressed or stable comprise or consist of X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.67 ’H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.17 ’H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51 .0413C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.55 13C ppm.

[0097] In certain embodiments, the method further comprises using one or more normalization NMR metabolite features in combination with the differential NMR metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

[0098] In certain embodiments, the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71.9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

[0099] In certain embodiments, the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.72 13C ppm, X59 having a chemical shift at 5.24 ’H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.83 ’H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.78 1H ppm and a chemical shift at 57.0813C ppm.

[0100] In certain embodiments, the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.

[0101] In another aspect, a system for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed or underimmunosuppressed using a computer implemented method, described herein, is provided, the system comprising: (a) a storage component for storing data, wherein the storage component has instructions for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed or under-immunosuppressed based on analysis of the NMR spectra of the one or more metabolites stored therein; (b) a computer processor programmed to analyze the NMR spectra using one or more algorithms, wherein the computer processor is coupled to the storage component and configured to execute the instructions stored in the storage component in order to receive the inputted NMR spectra, and analyze the NMR spectra according to a computer implemented method described herein; and (c) a display component for displaying information regarding whether the renal transplant recipient is over-immunosuppressed or under-immunosuppressed.

[0102] In another aspect, a non-transitory computer-readable medium comprising program instructions that, when executed by a processor in a computer, causes the processor to perform a computer implemented method, described herein, is provided.

[0103] In another aspect, a kit comprising the non-transitory computer-readable medium, described herein, and instructions for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed or under-immunosuppressed is provided.

[0104] In another aspect, a method of detecting under-immunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy is provided, the method comprising:(a) obtaining a urine sample from the renal transplant recipient; (b) measuring levels of one or more metabolites in the urine; and (c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is under-immunosuppressed.

[0105] In certain embodiments, the one or more metabolites are selected from glucuronic acid, glucaric acid, lactose, glucose-6-phosphate, 1 -methylhistamine, histidine, trigonelline, trimethylamine N-oxide (TMAO), hippuric acid, choline, phosphoethanolamine, phosphocholine, acetylglycine, phenyllactic acid, mycophenolic acid glucuronide (MPAG), creatine, and hydroxyphenylacetic acid.

[0106] In certain embodiments, the method further comprises altering the immunosuppressive therapy administered to the renal transplant recipient if the renal transplant recipient is determined to be under-immunosuppressed, wherein immunosuppression is increased if the renal transplant recipient is determined to be under-immunosuppressed.

[0107] In certain embodiments, the method further comprises performing diagnostic screening to determine whether the renal transplant recipient has graft dysfunction, subclinical graft rejection, or graft rejection if the renal transplant recipient is determined to be under-immunosuppressed; and treating the renal transplant recipient for the graft dysfunction, subclinical graft rejection, or graft rejection.

[0108] In certain embodiments, the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA) or a random forest (RF) machine learning algorithm.

[0109] In certain embodiments, the one or more metabolites are measured by a method comprising performing nuclear magnetic resonance (NMR) spectroscopy or mass spectrometry.

[0110] In certain embodiments, one or more differential NMR metabolite features are used to classify the renal transplant recipient as under-immunosuppressed or stable using the machine learning model, wherein the one or more differential NMR metabolite features are selected from X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X1223 having a chemical shift at 7.941H ppm and a chemical shift at 140.3013C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51 .04 13C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.

[0111] In certain embodiments, the differential NMR metabolite features used to classify the renal transplant recipient as under-immunosuppressed or stable comprise or consist of X65 having achemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.67 ’H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.17 ’H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51 .0413C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.55 13C ppm.

[0112] In another aspect, a method of monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy is provided, the method comprising: (a) obtaining a urine sample from the renal transplant recipient; and (b) measuring levels of one or more metabolites in the urine sample using nuclear magnetic resonance (NMR), wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, intensities of one or more differential NMR metabolite features of one or more metabolites in the urine sample are used to classify the kidney graft as stable or unstable at risk of injury using a first machine learning model, wherein the one or more differential NMR metabolite features are selected from X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm; and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, clinical information comprising gender of the renal transplant recipient, age of the renal transplant recipient at time of transplant of the kidney graft, plasma level of BK polyomavirus (BKV) in the renal transplant recipient, level of serum creatinine in the renal transplant recipient, and estimated glomerular filtration rate (eGFR) of the renal transplant recipient in combination with intensities of one or more differential NMR metabolite features of one or more metabolites in the urine sample are used to classify the kidney graft as stable or unstable at risk of injury using a second machine learning model, wherein the one or more differential NMR metabolite features are selected from an NMR metabolite feature having a chemical shift at 4.261H ppm and a chemical shift at 68.9913C ppm, an NMR metabolite feature having a chemical shift at 3.87 ’H ppm and a chemical shift at 58.813C ppm, an NMR metabolite feature having a chemical shift at 7.141H ppm and a chemical shift at 1 19.9313C ppm, an NMR metabolite feature having a chemical shift at 3.221H ppm and a chemical shift at 30.1213C ppm, an NMR metabolite feature having a chemical shift at 8.181H ppm and a chemical shift at 137.9213C ppm, and one or more NMR metabolite features for one or more metabolites selectedfrom 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3- dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine.

[0113] In certain embodiments, if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, the one or more differential NMR metabolite features comprise the X65 having the chemical shift at 4.061H ppm and the chemical shift at 85.9413C ppm, the X56 having the chemical shift at 4.67 ’H ppm and the chemical shift at 98.5713C ppm, the X360 having the chemical shift at 2.171H ppm and the chemical shift at 24.5113C ppm, the X333 having the chemical shift at 4.441H ppm and the chemical shift at 51 .0413C ppm, the X93 having the chemical shift at 4.081H ppm and the chemical shift at 74.8313C ppm, the X129 having the chemical shift at 3.281H ppm and the chemical shift at 62.2313C ppm, and the X161 having the chemical shift at 3.971H ppm and the chemical shift at 46.5513C ppm; and if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, the one or more differential NMR metabolite features comprise the NMR metabolite feature having the chemical shift at 4.261H ppm and the chemical shift at 68.9913C ppm, the NMR metabolite feature having the chemical shift at 3.871H ppm and the chemical shift at 58.8 13C ppm, the NMR metabolite feature having the chemical shift at 7.141H ppm and a chemical shift at 119.9313C ppm, the NMR metabolite feature havingthe chemical shift at 3.221H ppm and the chemical shift at 30.1213C ppm, the NMR metabolite feature having the chemical shift at 8.181H ppm and the chemical shift at 137.9213C ppm, and the one or more NMR metabolite features for the one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine.

[0114] In certain embodiments, wherein the kidney graft is classified as unstable at risk of injury, the method further comprises determining whether the renal transplant recipient is underimmunosuppressed or over-immunosuppressed, wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, clinical information comprising the age of the renal transplant recipient at the time of transplant of the kidney graft and the plasma level of BKV in the renal transplant recipient in combination with intensities of differential NMR metabolite features comprising X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm and X2473 having a chemicalshift at 3.81 ’H ppm and a chemical shift at 63.9313C ppm are used to classify the renal transplant recipient as under-immunosuppressed or over-immunosuppressed using a third machine learning model; and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, clinical information comprising the gender of the renal transplant recipient, the age of the renal transplant recipient at time of transplant of the kidney graft, the plasma level of BKV in the renal transplant recipient, the level of serum creatinine in the renal transplant recipient, and the eGFR of the renal transplant recipient in combination with intensities of the one or more differential NMR metabolite features selected from the NMR metabolite feature having the chemical shift at 4.261H ppm and the chemical shift at 68.9913C ppm, the NMR metabolite feature having the chemical shift at 3.871H ppm and the chemical shift at 58.813C ppm, the NMR metabolite feature having the chemical shift at 7.141H ppm and the chemical shift at 1 19.9313C ppm, the NMR metabolite feature having the chemical shift at 3.221H ppm and the chemical shift at 30.1213C ppm, the NMR metabolite feature having the chemical shift at 8.181H ppm and the chemical shift at 137.9213C ppm, and the one or more NMR metabolite features for the one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine are used to classify the renal transplant recipient as under-immunosuppressed or overimmunosuppressed using a fourth machine learning model.

[0115] In certain embodiments, the method further comprises altering the immunosuppressive therapy administered to the renal transplant recipient if the renal transplant recipient is determined to be under-immunosuppressed or over-immunosuppressed, wherein immunosuppression is increased if the renal transplant recipient is determined to be under-immunosuppressed and decreased if the renal transplant recipient is determined to be over-immunosuppressed.

[0116] In certain embodiments, the method further comprises performing diagnostic screening to determine whether the renal transplant recipient has an infection, a malignancy, or polyomavirus- associated nephropathy if the renal transplant recipient is determined to be overimmunosuppressed; and treating the renal transplant recipient for the infection, the malignancy, or the polyomavirus-associated nephropathy if the renal transplant recipient is diagnosed as having the infection, the malignancy, or the polyomavirus-associated nephropathy.

[0117] In certain embodiments, the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, the one or metabolites comprise glucuronic acid, trigonelline, trimethylamine N-oxide (TMAO), and hippuric acid.

[0118] In certain embodiments, the method further comprises performing diagnostic screening to determine whether the renal transplant recipient has graft dysfunction, subclinical graft rejection, or graft rejection if the renal transplant recipient is determined to be under-immunosuppressed; and treating the renal transplant recipient for the graft dysfunction, subclinical graft rejection, or graft rejection.

[0119] In certain embodiments, the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA) or a random forest (RF) machine learning algorithm.

[0120] In certain embodiments, the NMR spectroscopy is multi-dimensional NMR spectroscopy. In some embodiments, the NMR spectroscopy comprises one-dimensional (1 D)1H NMR spectroscopy, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectroscopy, or a combination thereof. In some embodiments, the NMR is performed with non-uniformed sampling (NUS).

[0121] In another aspect, a computer implemented method for monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy is provided, the computer performing steps comprising: (a) receiving nuclear magnetic resonance (NMR) spectra of one or more metabolites from a urine sample obtained from the renal transplant recipient; (b) measuring intensities of one or more differential NMR metabolite features in the NMR spectra, wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, the intensities of the one or more differential NMR metabolite features of the one or more metabolites in the urine sample are used to classify the kidney graft as stable or unstable at risk of injury using a first machine learning model, wherein the one or more differential NMR metabolite features are selected from X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.67 ’H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.17 ’H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51 .0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm; and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, clinical information comprising gender of the renal transplant recipient, age of the renal transplant recipient at time of transplant of the kidney graft, plasma level of BK polyomavirus (BKV) in the renal transplant recipient, level of serum creatinine in the renal transplant recipient, and estimated glomerular filtration rate (eGFR) of the renal transplant recipient in combination with theintensities of one or more differential NMR metabolite features of the one or more metabolites in the urine sample are used to classify the kidney graft as stable or unstable at risk of injury using a second machine learning model, wherein the one or more differential NMR metabolite features are selected from an NMR metabolite feature having a chemical shift at 4.261H ppm and a chemical shift at 68.99 13C ppm, an NMR metabolite feature having a chemical shift at 3.871H ppm and a chemical shift at 58.813C ppm, an NMR metabolite feature having a chemical shift at 7.141H ppm and a chemical shift at 1 19.9313C ppm, an NMR metabolite feature having a chemical shift at 3.221H ppm and a chemical shift at 30.1213C ppm, an NMR metabolite feature having a chemical shift at 8.181H ppm and a chemical shift at 137.9213C ppm, and one or more NMR metabolite features for one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine; and (c) displaying information regarding whether the kidney graft is stable or unstable at risk of injury.

[0122] In certain embodiments, wherein the kidney graft is classified as unstable at risk of injury, the method further comprises determining whether the renal transplant recipient is underimmunosuppressed or over-immunosuppressed, wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, clinical information comprising the age of the renal transplant recipient at the time of transplant of the kidney graft and the plasma level of BKV in the renal transplant recipient in combination with the intensities of differential NMR metabolite features comprising X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm and X2473 having a chemical shift at 3.81 ’H ppm and a chemical shift at 63.9313C ppm are used to classify the renal transplant recipient as under-immunosuppressed or over-immunosuppressed using a third machine learning model, and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, clinical information comprising the gender of the renal transplant recipient, the age of the renal transplant recipient at time of transplant of the kidney graft, the plasma level of BKV in the renal transplant recipient, the level of serum creatinine in the renal transplant recipient, and the eGFR of the renal transplant recipient in combination with the intensities of the one or more differential NMR metabolite features selected from the NMR metabolite feature having the chemical shift at 4.261H ppm and the chemical shift at 68.9913C ppm, the NMR metabolite feature having the chemical shift at 3.871H ppm and the chemical shift at 58.813C ppm, the NMR metabolite feature having the chemical shift at 7.141H ppm and the chemical shift at 1 19.9313Cppm, the NMR metabolite feature having the chemical shift at 3.221H ppm and the chemical shift at 30.1213C ppm, the NMR metabolite feature having the chemical shift at 8.181H ppm and the chemical shift at 137.9213C ppm, and the one or more NMR metabolite features for the one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine are used to classify the renal transplant recipient as under-immunosuppressed or overimmunosuppressed using a fourth machine learning model; and displaying information regarding whether the renal transplant recipient is under-immunosuppressed or over-immunosuppressed.

[0123] In certain embodiments, the method further comprises storing the information regarding whether the renal transplant recipient is under-immunosuppressed or over-immunosuppressed in a database.

[0124] In another aspect, a system for monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy using a computer implemented method, described herein, is provided, the system comprising: (a) a storage component for storing data, wherein the storage component has instructions for monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy based on analysis of the NMR spectra of the one or more metabolites stored therein; (b) a computer processor programmed to analyze the NMR spectra using one or more algorithms, wherein the computer processor is coupled to the storage component and configured to execute the instructions stored in the storage component in order to receive the inputted NMR spectra, and analyze the NMR spectra according to the computer implemented method described herein; and (c) a display component for displaying information regarding whether the renal transplant recipient is stable, unstable at risk of injury, underimmunosuppressed, or over-immunosuppressed.

[0125] In another aspect, a non-transitory computer-readable medium comprising program instructions that, when executed by a processor in a computer, causes the processor to perform a computer implemented method, described herein, is provided.

[0126] In another aspect, a kit comprising the non-transitory computer-readable medium, described herein, and instructions for monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy is provided.BRIEF DESCRIPTION OF THE DRAWINGS

[0127] FIG. 1. Delicate balance of immunosuppression. RTRs receive life-long immunosuppressants (IS), however it is a clinical challenge to determine the right dose for an individual. Too little IS risks under-immunosuppression and graft rejection, while too much risks overimmunosuppression and increased likelihood of infection and polyomavirus associated nephropathy (PVAN). The goal is to reach a “stable” state where risk of rejection and infection is minimized.

[0128] FIGS. 2A-2D. Urine metabolites in Stable and PVAN RTRs. (FIG. 2A) Representative 1 D NMR spectrum of a Stable sample. (FIG. 2B) Representative 1 D NMR spectrum of a RTR with PVAN indicating overimmunosuppression. Insets at top of each panel represent zoomed-in aliphatic and aromatic regions of the 1 D spectrum. (FIG. 2C) Representative 2D HSQC NMR spectrum from a Stable sample. (FIG. 2D) Representative 2D HSQC NMR spectrum from a RTR with PVAN indicating overimmunosuppression.

[0129] FIGS. 3A-3B. QA / QC of 1 H-13C HSQC metabolomics data for PVAN and Stable samples. (FIG. 3A) The overall resonance intensity sum (total sum), mean, median, maximum, number of peaks, and DSS intensity was compared for each sample. The majority of the samples had relatively consistent values, suggesting good quality data. Those with >2 or >1 .5 standard deviations from the mean were flagged in blue and green respectively. Those spectra were manually assessed and where appropriate samples were removed due to poor spectral quality and the potential to confound results. (FIG. 3B) For the N=100 samples that passed the QA / QC assessment, PCA was performed to further evaluate consistency in the data. The majority of the samples clustered together suggesting a similar composition of metabolites.

[0130] FIGS. 4A-4B. Differential NMR metabolite features in PVAN vs Stable. (FIG. 4A) Frequency of significantly different (p-value<0.05, FC>1 .5 and CV <100% per class) clusters when comparing PVAN vs Stable samples during the bootstrap KW analysis of 1000 independent tests. (FIG. 4B) Violin plots of the features that were significantly different in 70% of all KW analysis when comparing PVAN (red) to Stable samples (green). Each dot represent a unique sample.

[0131] FIGS. 5A-5B. Building an NMR ML model to distinguish PVAN from Stable. (FIG. 5A) As a first step to generate a classifier the data was randomly split (70 / 30) into training and validation sets, wherein each group was required to maintain consistent meta-data. In this way the both the training and validation data are well-representative of overall cohort. (FIG. 5B) There were no significant differences in patient characteristics comparing training and validation including: % Stable, age, sex, days post-transplant, % indicative biopsy, serum creatinine and eGFR. Metadata was unavailable from the following samples, sex (1 validation sample), serum creatinine (18 training, 5 validation), and eGFR (5 training, 1 validation).

[0132] FIG. 6. NMR Classifier selection. Using the KW features and KW-features refined by Boruta, the training data was split 70 / 30 into training and test set and 10-fold cross-fold validated ML models were constructed using ENET, OPLS-DA and RF and assessed based on accuracy, area under the curve (AUG), sensitivity and specificity in training and testing data. To mitigate overfitting, models that were significantly different between training and test (red boxes) were eliminated. The remaining three models demonstrating high performance to distinguish PVAN vs. Stable. Specifically, the OPLS-DA models with KW feature selection had the average cross-validated accuracy, AUG, sensitivity, and specificity of 85.4 ± 3.7%, 90.2 ± 3.0%, 58.3 ± 12.4%, 94.0 ± 1.8%. The OPLS-DA models with Boruta selection had the average cross-validated accuracy, AUG, sensitivity, and specificity of 86.0 ± 3.3%, 88.7 ± 4.5%, 58.3 ± 14.2%, 94.7 ± 2.1%. The RF model with Boruta feature selection had the average cross-validated accuracy, AUG, sensitivity, and specificity of 80.2 ± 4.6%, 78.1 ± 7.0%, 43.3 ± 14.1%, 91.8 ± 2.3%. The OPLS-DA model with KW features was selected as the top model based on AUC (green box).

[0133] FIGS. 7A-7B. myOLARIS-PVAN Score has high accuracy to classify PVAN from Stable. Waterfall plot for training (FIG. 7A) and validation (FIG. 7B) data with predicted PVAN scoring over 0.28 and predicted Stable scoring under 0.28. T rue PVAN are colored red and true Stable are colored green. Results from the receiver operator characteristic (ROC) analysis are provided.

[0134] FIGS. 8A-8C. Annotation of Differential Metabolites in myOLARIS-PVAN model. (FIG. 8A) Summary of top features in the ML model. (FIG. 8B) Spike in confirmed X109 as lysine wherein the intensity of the differential peak (red) increased in intensity when a lysine standard was spiked to the sample. (FIG. 8C) X32, X40, X44 and X205 are well resolved peaks but did not match to any known compounds in the library. X1103 matched to several putative sugars, which it is not easily distinguished by NMR or MS. Efforts are underway to determine the identity of the unknowns. For NMR models these features can be monitored by their unique chemical shift.

[0135] FIGS. 9A-9B. Refined myOLARIS-PVAN Score maintains accuracy to classify PVAN from Stable. (FIG. 9A) Waterfall plot for training (left) and validation (right) data after removing X1 103 and updated cut off. Predicted PVAN scoring over 0.30 and predicted Stable scoring under 0.30. True PVAN are colored red and true Stable are colored green. Results from the receiver operator characteristic (ROC) analysis are provided. (FIG. 9B) Stress test where each feature was removed to evaluate performance of the model (decrease in blue and increase in red). Removing any feature significantly reduced AUC and sensitivity.

[0136] FIG. 10. Identifying a defined set of normalization factors. Correlation between the normalization factors using PQN and LASSO with 10 defined features.

[0137] FIGS. 11A-11 B. Champion myOLARIS-PVAN Score demonstrates high accuracy to classify PVAN from Stable. (FIG. 11 A) Waterfall plot for training (left) and validation (right) using defined features (5 differential and 10 NFs). Predicted PVAN scoring over 0.42 and predicted Stable scoring under 0.42. True PVAN are colored red and true Stable are colored green. Results from the receiver operator characteristic (ROC) analysis are provided. (FIG. 11 B) Stress test where each feature was removed to evaluate performance of the model (decrease in blue and increase in red). Removing any feature significantly reduced accuracy, AUG and specificity.

[0138] FIGS. 12A-12H. Annotation of Differential and Normalization Factors in champion myOLARISPVAN model. (FIG. 12A) Summary of all features in the champion ML model. Spike in confirmation of (FIG. 12B) hippuric acid (FIG. 12C) lysine, (FIG. 12D) MPAG, (FIG. 12E) mannitol (FIG. 12F) TMAO and (FIG. 12G) glutamine. (FIG. 12H) X280 is in the fatty acid part of the spectra and X205, X108, X48 did not match to any molecules in the library. Efforts are underway to determine the identity of the unknowns.

[0139] FIGS. 13A-13D. QA / QC of LC-MS metabolomics data for PVAN and Stable samples. (FIG. 13A) The overall intensity sum (total sum), mean, median, and maximum were compared for each sample. The majority of the samples had relatively consistent values, suggesting good quality data. Those with >2 or >1.5 standard deviations from the mean were flagged in blue and green respectively. Those spectra were manually assessed and based on this retained for downstream analysis. (FIG. 13B) For the N=41 samples that passed the QA / QC check and 16 MSQC2 samples, PCA was performed to further evaluate consistency in the data. The majority of the samples overlapped suggesting a similar composition of metabolites. The pooled QC (in yellow) samples are in the middles of all urine samples. (FIG. 13C) The total sum quality metric plot visualizes the total signal intensity for each sample along the LCMS sequence. The trend of the pooled QC (in red) samples forms a horizontal linear line, demonstrating the consistency of LC-MS performance. (FIG. 13D) Histogram of %CV for metabolomic features in the pooled QC samples. The majority of the features have a %CV of less than 25%. Three features, Flavin adenine dinucleotide [pos] - Tier 2, Dihydroxyacetone phosphate [neg] - Tier 2, and UDP-N-acetyl-alpha-D-glucosamine [pos], exhibit a %CV exceeding 50%.

[0140] FIGS. 14A-14B. Differential Metabolites in the LC-MS myOLARIS-PVAN model. (FIG. 14A) Violin plots and (FIG. 14B) statistical summary for 3 differential metabolites identified via LC-MS including fold-change (FC) and p-value.

[0141] FIG. 15. Confirmation of hit identities using authentic standards. LC-MS of benzoyl formic acid ([M-H]-, m / z 149.09; SRMs: m / z 104.91 and m / z 77.07), cytidine ([M+H]+, m / z 244.09; SRM:m / z 11 1.90), and caffeine ([M+H]+, m / z 195.00; SRMs: m / z 137.90 and m / z 1 10.00 ) in the pooled QC sample (top) versus authentic standards (bottom).

[0142] FIGS. 16A-16C. myOLARIS-PVAN LCMS Score. (FIG. 16A) 10-fold CV performance boxplot for LCMS. Training performances are in orange and Testing performance are in blue. (FIG. 16B) Waterfall plot for training samples, with predicted PVAN scoring over 0.33 and predicted Stable scoring under 0.33. True PVAN are colored red and true Stable are colored green. Results from the receiver operator characteristic (ROC) analysis are provided. (FIG. 16C) Stress test where each feature was removed to evaluate performance of the model (decrease in blue and increase in red). Removing any feature significantly reduced AUC.

[0143] FIGS. 17A-17B. Comparison between NMR and MS. (FIG. 17A) Summary table of all identified metabolites in either the NMR or MS myOLARIS-PVAN models, with FC, p-value and %CV described for matched samples. The correlation between raw intensity for each metabolite is provided from a SG corrected urine analysis with (FIG. 17B) scatter plots of NMR intensity and MS peak area values for the same metabolites between platforms.

[0144] FIGS. 18A-18X. Levels measured by NMR spectroscopy of metabolite resonances at 1 .991 + / - .25 ppm x 40.132 + / - 0.45 ppm (FIG. 18A), 2.566 + / - .25 ppm x 47.724 + / - 0.45 ppm (FIG. 18B), 2.712 + / - .25 ppm x 47.752 + / - 0.45 ppm (FIG. 18C), 3.787 + / - .25 ppm x 73.579 + / - 0.45 ppm (FIG. 18D), 3.813 + / - .25 ppm x 62.593 + / - 0.45 ppm (FIG. 18E), 3.876 + / - .25 ppm x 35.932 + / - 0.45 ppm (FIG. 18F), 3.969 + / - .25 ppm x 46.487 + / - 0.45 ppm (FIG. 18G), 4.44 + / - .25 ppm x 50.942 + / - 0.45 ppm (FIG. 18H), 6.914 + / - .25 ppm x 115.425 + / - 0.45 ppm (FIG. 181), 6.974 + / - .25 ppm x 120.636 + / - 0.45 ppm (FIG. 18J), 7.085 + / - .25 ppm x 121.7 + / - 0.45 ppm (FIG. 18K), 7.274 + / - .25 ppm x116.531 + / - 0.45 ppm (FIG. 18L), 7.294 + / - .25 ppm x 132.721 + / - 0.45 ppm (FIG. 18M), 7.346 + / - .25 ppm x 121.584 + / - 0.45 ppm (FIG. 18N), 7.531 + / - .25 ppm x 129.748 + / - 0.45 ppm (FIG. 180),7.531 + / - .25 ppm x 131.363 + / - 0.45 ppm (FIG. 18P), 7.532 + / - .25 ppm x 134.804 + / - 0.45 ppm (FIG. 18Q), 7.618 + / - .25 ppm x 134.808 + / - 0.45 ppm (FIG. 18R), 7.624 + / - .25 ppm x 131.123 + / - 0.45 ppm (FIG. 18S), 7.817 + / - .25 ppm x 131.402 + / - 0.45 ppm (FIG. 18T), 7.818 + / - .25 ppm x 129.762 + / - 0.45 ppm (FIG. 18U), 8.08 + / - .25 ppm x 130.252 + / - 0.45 ppm (FIG. 18V), 8.841 + / - .25 ppm x 147.236 + / - 0.45 ppm (FIG. 18W), 9.117 + / - .25 ppm x 148.323 + / - 0.45 ppm (FIG. 18X) in the1H and13C dimensions respectively that were significantly different between altered ISR patients and control patients.

[0145] FIG. 19. Receiver operator curve (ROC) for the machine learning algorithm based on differential metabolites which produced a biomarker of response (BoR) score that differentiates altered ISR subjects from healthy controls with 81 .1% cvAUC.

[0146] FIGS. 20A-20C. Metabolite levels measured by NMR spectroscopy for levels of metabolite resonances at 2.792 + / - .25 ppm x 40.011 + / - .45 ppm (FIG. 20A), 3.714 + / - .25 ppm x 72.206 + / - .45 ppm (FIG. 20B), 3.009 + / - .25 ppm x 32.551 + / - .45 ppm (FIG. 20C) in the1H and13C dimensions respectively that were significantly different in kidney transplant subjects who were underimmunosuppressed compared to control subjects.

[0147] FIG. 21. Receiver operator curve (ROC) for the machine learning algorithm based on differential metabolites which produce a biomarker of response (BoR) score that differentiates kidney transplant subjects who were under-immunosuppressed from controls with 87.1% cvAUC.

[0148] FIGS. 22A-22R. Metabolite levels measured by NMR spectroscopy for levels of metabolite resonances at 2.512 + / - .25 ppm x 28.032 + / - .45 ppm (FIG. 22A), 2.787 + / - .25 ppm x 40.035 + / - .45 ppm (FIG. 22B), 3.01 + / - .25 ppm x 32.525 + / - .45 ppm (FIG. 22C), 3.637 + / - .25 ppm x 78.911 + / - .45 ppm (FIG. 22D), 3.714 + / - .25 ppm x 72.229 + / - .45 ppm (FIG. 22E), 3.722 + / - .25 ppm x 75.654 + / - .45 ppm (FIG. 22F), 3.851 + / - .25 ppm x 64.395 + / - .45 ppm (FIG. 22G), 3.966 + / - .25 ppm x 46.482 + / - .45 ppm (FIG. 22H), 5.016 + / - .25 ppm x 74.051 + / - .45 ppm (FIG. 221), 6.914 + / - .25 ppm x 1 15.449 + / - .45 ppm (FIG. 22J), 6.974 + / - .25 ppm x 120.651 + / - .45 ppm (FIG. 22K), 7.088 + / - .25 ppm x 121.707 + / - .45 ppm (FIG. 22L), 7.276 + / - .25 ppm x 116.555 + / - .45 ppm (FIG. 22M), 7.533 + / - .25 ppm x 131 .372 + / - .45 ppm (FIG. 22N), 7.534 + / - .25 ppm x 129.772 + / - .45 ppm (FIG. 220), 7.534 + / - .25 ppm x 134.842 + / - .45 ppm (FIG. 22P), 7.619 + / - .25 ppm x 134.811 + / - .45 ppm (FIG. 22Q), 7.817 + / - .25 ppm x 131.398 + / - .45 ppm (FIG. 22R) in the1H and13C dimensions respectively that were significantly different between under and over-immunosuppressed subjects.

[0149] FIG. 23. Receiver operator curve (ROC) for the machine learning algorithm based on differential metabolites which produce a biomarker of response (BoR) score that differentiates between under and over-immunosuppressed subjects with 90.9% cvAUC.

[0150] FIGS. 24A-24L. Metabolite levels measured by NMR spectroscopy for levels of metabolite resonances at 2.71 1 + / - .25 ppm x 47.702 + / - .45 ppm (FIG. 24A), 3.969 + / - .25 ppm x 46.484 + / - .45 ppm (FIG. 24B), 7.086 + / - .25 ppm x 121 .698 + / - .45 ppm (FIG. 24C), 7.274 + / - .25 ppm x 116.53 + / - .45 ppm (FIG. 24D), 7.347 + / - .25 ppm x 121.578 + / - .45 ppm (FIG. 24E), 7.532 + / - .25 ppm x 129.75 + / - .45 ppm (FIG. 24F), 7.532 + / - .25 ppm x 131.359 + / - .45 ppm (FIG. 24G), 7.533 + / - .25 ppm x 134.812 + / - .45 ppm (FIG. 24H), 7.618 + / - .25 ppm x 134.804 + / - .45 ppm (FIG. 241), 7.817 + / - .25 ppm x 131.389 + / - .45 ppm (FIG. 24J), 7.818 + / - .25 ppm x 129.746 + / - .45 ppm (FIG. 24K), 9.1 17 + / - .25 ppm x 148.321 + / - .45 ppm (FIG. 24L) in the1H and13C dimensions respectively that were significantly different between subjects who had appropriate ISR compared to those that did not.

[0151] FIG. 25. Receiver operator curve (ROC) for the machine learning algorithm based on differential metabolites which produce a biomarker of response (BoR) score that differentiates subjects who had appropriate ISR compared to those that did not with 75% cvAUC.

[0152] FIGS. 26A-26P. Metabolite levels measured by NMR spectroscopy for levels of metabolite resonances at 2.19 + / - .25 ppm x 24.54 + / - .45 ppm (FIG. 26A), 2.22 + / - .25 ppm x 39.75 + / - .45 ppm (FIG. 26B), 2.82 + / - .25 ppm x 30 + / - .45 ppm (FIG. 26C), 2.89 + / - .25 ppm x 32.96 + / - .45 ppm (FIG. 26D), 3.12 + / - .25 ppm x 32.81 + / - .45 ppm (FIG. 26E), 3.38 + / - .25 ppm x 76.2 + / - .45 ppm (FIG. 26F), 3.39 + / - .25 ppm x 76.25 + / - .45 ppm (FIG. 26G), 3.62 + / - .25 ppm x 78.2 + / - .45 ppm (FIG. 26H), 3.75 + / - .25 ppm x 62 + / - .45 ppm (FIG. 26I), 3.97 + / - .25 ppm x 46.51 + / - .45 ppm (FIG. 26J), 4.05 + / - .25 ppm x 58.5 + / - .45 ppm (FIG. 26K), 4.45 + / - .25 ppm x 50.9 + / - .45 ppm (FIG. 26L), 7.534 + / - .25 ppm x 131 .3 + / - .45 ppm (FIG. 26M), 7.619 + / - .25 ppm x 134.8 + / - .45 ppm (FIG. 26N), 7.62 + / - .25 ppm x 131 .1 + / - .45 ppm (FIG. 260), 7.82 + / - .25 ppm x 129.7 + / - .45 ppm (FIG. 26P) in the1H and13C dimensions respectively that were significantly different between subjects who developed BKVIN and control subjects.

[0153] FIG. 27. Receiver operator curve (ROC) for the machine learning algorithm based on differential metabolites which produce a biomarker of response (BoR) score that differentiates subjects who developed BKVIN and control subjects with 91 .5% cvAUC.

[0154] FIGS. 28A-28O. Metabolite levels measured by NMR spectroscopy for levels of metabolite resonances at 2.2 + / - .25 ppm x 39.75 + / - .45 ppm (FIG. 28A), 2.82 + / - .25 ppm x 30 + / - .45 ppm (FIG. 28B), 3.71 + / - .25 ppm x 72.1 + / - .45 ppm (FIG. 28C), 3.75 + / - .25 ppm x 62 + / - .45 ppm (FIG. 28D), 3.973 + / - .25 ppm x 46.56 + / - .45 ppm (FIG. 28E), 4.05 + / - .25 ppm x 58.6 + / - .45 ppm (FIG. 28F), 4.45 + / - .25 ppm x 50.8 + / - .45 ppm (FIG. 28G), 7.5 + / - .25 ppm x 129.7 + / - .45 ppm (FIG. 28H), 7.533 + / - .25 ppm x 134.8 + / - .45 ppm (FIG. 281), 7.534 + / - .25 ppm x 131.3 + / - .45 ppm (FIG. 28J), 7.618 + / - .25 ppm x 134.7 + / - .45 ppm (FIG. 28K), 7.62 + / - .25 ppm x 131.1 + / - .45 ppm (FIG. 28L), 7.8 + / - .25 ppm x 131 .4 + / - .45 ppm (FIG. 28M), 7.82 + / - .25 ppm x 129.7 + / - .45 ppm (FIG. 28N), 9.11 + / - .25 ppm x 148.25 + / - .45 ppm (FIG. 280) in the1H and13C dimensions respectively that were significantly different between male over-immunosuppressed subjects and male control subjects.

[0155] FIG. 29. Receiver operator curve (ROC) for the machine learning algorithm based on differential metabolites which produce a biomarker of response (BoR) score that differentiates male over-immunosuppressed subjects and male control subjects with 100% cvAUC.

[0156] FIGS. 30A-30M. Metabolite levels measured by NMR spectroscopy for levels of metabolite resonances at 1.27 + / - .25 ppm x 30.8 + / - .45 ppm (FIG. 30A), 1.91 + / - .25 ppm x 32.6 + / - .45 ppm (FIG. 30B), 2.22 + / - .25 ppm x 39.75 + / - .45 ppm (FIG. 30C), 2.51 + / - .25 ppm x 28.05 + / - .45 ppm(FIG. 30D), 3.12 + / - .25 ppm x 32.76 + / - .45 ppm (FIG. 30E), 3.163 + / - .25 ppm x 44.09 + / - .45 ppm (FIG. 30F), 3.25 + / - .25 ppm x 30.33 + / - .45 ppm (FIG. 30G), 3.38 + / - .25 ppm x 76.2 + / - .45 ppm (FIG. 30H), 3.6 + / - .25 ppm x 73.11 + / - .45 ppm (FIG. 301), 3.72 + / - .25 ppm x 69.98 + / - .45 ppm (FIG. 30J), 7.16 + / - .25 ppm x 122.3 + / - .45 ppm (FIG. 30K), 7.48 + / - .25 ppm x 114.6 + / - .45 ppm (FIG. 30L), 7.65 + / - .25 ppm x 119.8 + / - .45 ppm (FIG. 30M) in the1H and13C dimensions respectively that were significantly different between female over-immunosuppressed subjects and female control subjects.

[0157] FIG. 31. receiver operator curve (ROC) for the machine learning algorithm based on differential metabolites which produce a biomarker of response (BoR) score that differentiates female over-immunosuppressed subjects and female control subjects with 100% cvAUC.

[0158] FIG. 32. High rates of hospitalization rates in first year post transplant led to poor health and economic outcomes. Despite routine lab testing, the overwhelming majority of transplant patients will be re-admitted to the hospital in the 1 st year (-63%). Those patients readmitted have increased risk of graft failure and death. Further the cost of re-admittance creates a significant burden to healthcare system, representing 20% of all Medicare payments for transplant. References:1Julien Hogan, et al. (2019) Transplantation Direct,2USRDS,3Kinza Iqbal, et al (2022) Frontiers in Medicine.

[0159] FIG. 33. Need for improved biomarkers to guide clinical decisions. After evaluation of lab testing, physicians first determine if labs are “off” and suggestive of an underlying problem such as infection, rejection or other complication. If so, the physician can order a biopsy, which remains the gold-standard to diagnose graft injury or they can adjust the levels of immunosuppressants (IS) or other medications. If the labs are deemed “ok” the physician can decide to maintain or reduce levels of IS to reduce complications associated with long-term IS usage. Aside from serum creatinine, which is considering a lagging indicator for graft injury, there are limited clinically validated biomarkers to guide clinical decision making to either order a biopsy or adjust dosing.

[0160] FIGS. 34A-34B. Serum creatinine fails to detect graft injury. (FIG. 34A) Waterfall plot of serum creatinine level prediction for the training samples with predicted Injury scored over 1 .895 and predicted Stable scoring under the line. T ure Injury are colored dark gray and true Stable are colored light gray. (FIG. 34B) Receiver operator characteristic (ROC) analysis with 0.654 AUG, 66.0% accuracy (ACC), 54.9% sensitivity (SEN) and 76.9% specificity (SPE). These results demonstrate shortcomings of serum creatinine to identify graft injury and / or guide clinical decision making.

[0161] FIG. 35. Sample collection map. Urine samples were collected from N=103 patients before a biopsy (protocol or indicative) 6 to 129 days post-transplant. Patients with biopsy proven Stable grafts (no histological signs of rejection or infection) are shown in dark gray and those with biopsy proven Injury (subclinical or clinical rejection and PVAN) are shown in light gray.

[0162] FIGS. 36A-36C. Differential metabolite resonances between Injury and Stable samples. (FIG. 36A) Volcano plot and (FIG. 36B) violin plots highlighting the significantly different metabolite resonances between Injury (dark gray) and Stable (light gray) samples. Using a KW-test of significance 8 significant features were identified with p-value <0.05 and an absolute FC difference of greater than 1.5, with 7 of the 8 also passing an FDR-adjusted p-value of 0.05. (FIG. 36C) Heatmap with unsupervised hierarchical clustering of differential features between Injury and Stable samples. Metabolite resonance intensity is colored with higher intensity in red and lower intensity in blue.

[0163] FIG. 37. Building an Injury classifier. The training data was split 70 / 30 into training and test sets, 10x cross-validated using 2 predefined feature sets and 3 machine learning models. Models with light gray squares were excluded due to potential overfitting. The top model (dark gray) was selected based on highest testing AUG.

[0164] FIGS. 38A-38B. myOLARIS Score Differentiates Injury from Stable. (FIG. 38A) Waterfall plot of myOLARIS score for the full training cohort with predicted Injury scored over 0.454 and predicted Stable scoring under the line. Ture Injury are colored dark gray and true Stable are colored light gray. (FIG. 38B) Receiver operator characteristic (ROC) analysis with 0.817 AUG, 79.3% accuracy (ACC), 78% sensitivity and 80.9% specificity at the 0.454 cut-off.

[0165] FIG. 39. Investigating incorrect predictions. We sought to determine any systematic bias or opportunities to improve the algorithm by examining the incorrect predictions for Stable and Injury which were classified as Tier 1 and Tier 2 based on distance from the cut-off to correct predictions.

[0166] FIGS. 40A-40B. Tacrolimus dose as a feature does not improve model performance. We included tacrolimus dosing information (TacDaily) as a feature in the ML and did not observe increased performance. (FIG. 40A) Waterfall plot of myOLARIS score. (FIG. 40B) ROC analysis with 0.819 AUC.

[0167] FIGS. 41A-41 B. Comparison of features significantly altered between correct and incorrect predictions. (FIG. 41 A) Venn diagram comparing differential features for Injury Tier 1 (I-T1 ), Injury Tier 2 (I-T2), Stable Tier 1 (S_T1 ) and Stable Tier 2 (S_T2). X1223 was significantly different all incorrect Injury and Stable Tier 1 predictions, X140 and X143 in all Injury incorrect predictions and X1 13 X104 X126 X114 in Injury Tier 2 and Stable Tier 1 . (FIG. 41 B) Table listing number of elements in each tier.

[0168] FIGS. 42A-42B. Champion model to differentiate Injury from Stable. (FIG. 42A) Waterfall plot of myOLARIS-KTdx score for the full training cohort with predicted Injury scored over 0.534 and predicted Stable scoring under the line. True Injury are colored purple and true Stable are coloredgreen. (FIG. 42B) Receiver operator characteristic (ROC) analysis with 0.837 AUC, 80.7% accuracy (ACC), 79.1% sensitivity and 82.2% specificity.

[0169] FIGS. 43A-43D. Validation performance of myOLARIS-KTdx to differentiate Injury from Stable. (FIG. 43A) Waterfall plot of myOLARIS-KTdx score for the validation cohort with predicted Injury scored over 0.534 and predicted Stable scoring under the line. True Injury are colored purple and true Stable are colored green. (FIG. 43B) Receiver operator characteristic (ROC) analysis with 0.922 AUC, 79.5% accuracy (ACC), 77.3% sensitivity (SEN) and 82.4% specificity (SPE). (FIG. 43C) Waterfall plot of serum creatinine level prediction for the same validation samples with predicted Injury scored over 1.895 and predicted Stable scoring under the line. Ture Injury are colored purple and true Stable are colored green. (FIG. 43D) Receiver operator characteristic (ROC) analysis with 0.592 AUC, 45.5% accuracy (ACC), 26.1% sensitivity (SEN) and 66.7% specificity (SPE). These results demonstrate superior ability of myOLARIS-KTdx over serum creatinine to detect graft injury.

[0170] FIGS. 44A-44B. Kit for monitoring a renal transplant recipient undergoing treatment with immunosuppressive therapy. FIG. 44A shows the contents of the kit, including a urine collection container, a biohazard bag, an insulated container for shipping a urine specimen to a diagnostic lab for testing, and instructions for using the kit. FIG. 44B shows the outer packaging of the kit.

[0171] FIG. 45. Patient sample graph. Urine samples (N=103) were collected from patients before a biopsy (protocol or indicative) 6 to 164 days post-transplant. Patients with biopsy proven Stable grafts (no histological signs of rejection or infection) are shown in light gray and those with biopsy proven Injury (subclinical or clinical rejection and PVAN) are shown in dark gray.

[0172] FIGS. 46A-46B. Serum creatinine fails to detect graft injury. FIG. 46A shows serum creatinine waterfall plot. FIG. 46B shows serum creatinine ROC plot.

[0173] FIG. 47. Study Overview schematic.

[0174] FIGS. 48A-48D. Differential metabolite resonances between Injury and Stable samples that were used for building an Injury classifier. FIG. 48A shows a Volcano Plot from KW results. FIG. 48B shows a Heatmap of the 7 differential features. FIG. 48C shows boxplots of training / testing cross- validation model performance. FIG. 48D shows a table summarizing the training model performance statistics.

[0175] FIGS. 49A-49B. Validation cohort independent of training machine learning model. FIG. 49A shows serum creatinine waterfall plot for validation cohort. FIG. 49B shows serum creatinine ROC plot for validation cohort. Serum creatinine for the validation cohort had far inferior discriminatory power compared to our model based on differential metabolite resonances.

[0176] FIGS. 50A-50B. Our model demonstrates superior performance to differentiate Injury from Stable in validation cohort. FIG. 50A shows a waterfall plot for the validation cohort. FIG. 50B shows a ROC Plot for the validation cohort.

[0177] FIGS. 51A-51 I. Our model has superior ability to detect graft injury. FIG. 51 A shows serum creatinine waterfall plot for the training cohort. FIG. 51 B shows serum creatinine ROC plot for the training cohort. FIG. 51C shows1H-13C HSQC NMR metabolite signature. FIG. 51 D shows NMR metabolite signature waterfall plot for the training cohort. FIG. 51 E shows NMR metabolite signature ROC plot for the training cohort. FIG. 51 F shows serum creatinine waterfall plot for the validation cohort. FIG. 51 G shows serum creatinine ROC plot for the validation cohort. FIG. 51 H shows NMR metabolite signature waterfall plot for the validation cohort. FIG. 511 shows NMR metabolite signature ROC plot for the validation cohort.

[0178] FIGS. 52A-52D. Differentiating Under from Over-immunosuppression. FIG. 52A shows waterfall plot for training cohort true injury. FIG. 52B shows ROC plot for training cohort true injury. FIG. 52C shows waterfall plot for validation cohort true injury. FIG. 52D shows ROC plot for validation cohort true injury.

[0179] FIG. 53. 2-step prediction strategy.

[0180] FIG. 54. Patient case studies in a surveillance setting and an indication setting.

[0181] FIGS. 55A-55K. Expanded kidney transplant (KT) follow up study (PD180) shows that myOLARIS-KTdx has superior ability to detect graft injury compared to serum creatinine. (FIG. 55A) Patient Sample graph. Urine samples (N=273) were collected from patients before a biopsy 186 to 2953 days post-transplant (up to 8 years) Patients with biopsy proven Stable grafts (no histological signs of rejection or infection) are shown in light gray and those with biopsy proven Injury (subclinical or clinical rejection and PVAN) are shown in dark gray. (FIG. 55B) Serum creatinine waterfall plot for training cohort. (FIG. 55C) Serum creatinine ROC plot for training cohort. (FIG. 55D) myOLARIS- KTdx waterfall plot fortraining cohort. (FIG. 55E) myOLARIS-KTdx ROC plot for training cohort. (FIG. 56F) Prevalence of NPV and PPV for training cohort. FIG. 55G) Serum creatinine waterfall plot for validation cohort. (FIG. 55H) Serum creatinine ROC plot for validation cohort. (FIG. 551) myOLARIS- KTdx waterfall plot for validation cohort. (FIG. 55J) myOLARIS-KTdx ROC plot for validation cohort. (FIG. 55K) Prevalence of NPV and PPV for validation cohort.

[0182] FIGS. 56A-56D. PD180 Differentiating Under from Over-immunosuppression. (FIG. 56A) Training Cohort True Injury waterfall plot. (FIG. 56B) Training Cohort True Injury ROC plot. (FIG. 56C) Validation Cohort True Injury waterfall plot. (FIG. 56D) Validation Cohort True Injury ROC plot. (FIG. 56E) Prevalence of NPV and PPV.

[0183] FIG. 57. Dual strategy for monitoring kidney transplant patients <180 days post-transplant and patients >180 days post-transplant.

[0184] FIG. 58. Performance of myOLARIS-KTdx across all samples to date. Samples were collected 6-2953 days post-transplant (6 days to 8 years post-transplant). All samples were matched with a biopsy. Retrospective samples from Seoul National University Hospital Korea and University of Leuven Belgium have also been analyzed.

[0185] FIG. 59. myOLARIS-KTdx maps changes in graft status across time.DETAILED DESCRIPTION OF THE INVENTION

[0186] Compositions, methods, kits, systems, and software are provided for detecting overimmunosuppression or under-immunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy. In particular, the methods utilize a urine-based metabolite signature that differentiates over-immunosuppressed or under-immunosuppressed renal transplant recipients from those with a stable graft with high accuracy. In some embodiments, nuclear magnetic resonance spectroscopy and / or liquid chromatography-mass spectrometry techniques are used to detect metabolite features that differentiate patients who are over-immunosuppressed or under-immunosuppressed from those with stable grafts.

[0187] Before the present compositions, methods, kits, systems, and software are described, it is to be understood that this invention is not limited to particular methods or compositions described, as such may, of course, vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting, since the scope of the present invention will be limited only by the appended claims.

[0188] Where a range of values is provided, it is understood that each intervening value, to the tenth of the unit of the lower limit unless the context clearly dictates otherwise, between the upper and lower limits of that range is also specifically disclosed. Each smaller range between any stated value or intervening value in a stated range and any other stated or intervening value in that stated range is encompassed within the invention. The upper and lower limits of these smaller ranges may independently be included or excluded in the range, and each range where either, neither or both limits are included in the smaller ranges is also encompassed within the invention, subject to any specifically excluded limit in the stated range. Where the stated range includes one or both of the limits, ranges excluding either or both of those included limits are also included in the invention.

[0189] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Although any methods and materials similar or equivalent to those described herein can be used inthe practice or testing of the present invention, some potential and preferred methods and materials are now described. All publications mentioned herein are incorporated herein by reference to disclose and describe the methods and / or materials in connection with which the publications are cited. It is understood that the present disclosure supersedes any disclosure of an incorporated publication to the extent there is a contradiction.

[0190] As will be apparent to those of skill in the art upon reading this disclosure, each of the individual embodiments described and illustrated herein has discrete components and features which may be readily separated from or combined with the features of any of the other several embodiments without departing from the scope or spirit of the present invention. Any recited method can be carried out in the order of events recited or in any other order which is logically possible.

[0191] It must be noted that as used herein and in the appended claims, the singular forms "a", "an", and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to "a biomarker" includes a plurality of such biomarkers and reference to "the metabolite" includes reference to one or more metabolites and equivalents thereof, known to those skilled in the art, and so forth.

[0192] The publications discussed herein are provided solely for their disclosure prior to the filing date of the present application. Nothing herein is to be construed as an admission that the present invention is not entitled to antedate such publication by virtue of prior invention. Further, the dates of publication provided may be different from the actual publication dates which may need to be independently confirmed.Definitions

[0193] Biomarkers. The term “biomarker” as used herein refers to a compound, such as a metabolite or a metabolic byproduct which is present at different concentrations, levels, or frequencies in one urine sample compared to another, such as a urine sample from a renal transplant recipient who is over-immunosuppressed or under-immunosuppressed compared to a renal transplant recipient who is stable (e.g., an individual known to not have over-immunosuppression, underimmunosuppression, graft dysfunction, subclinical graft rejection, graft rejection, or an infection).

[0194] A "reference level" or "reference value" of a biomarker means a level of the biomarker that is indicative of a particular state. For example, a reference level of a biomarker in a urine sample may correlate with over-immunosuppression, under-immunosuppression, graft dysfunction, subclinical graft rejection, graft rejection, an infection, or a stable transplant. A "reference level" of a biomarker may be an absolute or relative amount or concentration of the biomarker, a presence or absence of the biomarker, a range of amount or concentration of the biomarker, a minimum and / or maximumamount or concentration of the biomarker, a mean amount or concentration of the biomarker, and / or a median amount or concentration of the biomarker; and, in addition, "reference levels" of combinations of biomarkers may also be ratios of absolute or relative amounts or concentrations of two or more biomarkers with respect to each other. Appropriate reference levels of biomarkers for a particular state or lack thereof may be determined by measuring levels of desired biomarkers in urine samples for one or more renal transplant recipients, and such reference levels may be tailored to specific populations of subjects (e.g., a reference level may be age-matched or gender-matched so that comparisons may be made between biomarker levels in urine samples for renal transplant recipients of a certain age or gender. Such reference levels may also be tailored to specific techniques that are used to measure levels of biomarkers in urine samples (e.g., NMR, LC-MS, etc.), where the levels of biomarkers may differ based on the specific technique that is used.

[0195] A "similarity value" is a number that represents the degree of similarity between two things being compared. For example, a similarity value may be a number that indicates the overall similarity between a urine biomarker profile using specific biomarkers and reference value ranges for the biomarkers in one or more control urine samples or a reference profile (e.g., the similarity to a "stable transplant" biomarker profile, “over-immunosuppression” biomarker profile, “underimmunosuppression” biomarker profile, “PVAN” biomarker profile, "infection” biomarker profile, "subclinical graft rejection” biomarker profile, or "graft rejection" biomarker profile). The similarity value may be expressed as a similarity metric, such as a correlation coefficient, or may simply be expressed as the difference in an amount of a biomarker, or the aggregate of differences in amounts of biomarkers in a urine sample and a control urine sample or reference profile.

[0196] The terms "quantity", "amount", and "level" are used interchangeably herein and may refer to an absolute quantification of a molecule or an analyte in a urine sample, or to a relative quantification of a molecule or analyte in a urine sample, i.e., relative to another value such as relative to a reference value as taught herein, or to a range of values for a biomarker. These values or ranges can be obtained from urine samples from a single subject or a group of subjects.

[0197] The term “urine sample” with respect to a renal transplant recipient encompasses samples of urine taken before, during, or after treatment with immunosuppressive therapy (e.g., to prevent transplant rejection). Urine samples can be obtained by any suitable method such as by having the subject urinate into a collection container or by collection of urine using a urinary catheter. The definition also includes urine samples that have been manipulated in any way after their procurement, such as by treatment with reagents, washed, or enriched for particular types of molecules, e.g., metabolite biomarkers.

[0198] The term “assaying” is used herein to include the physical steps of manipulating a urine sample to generate data related to the urine sample. As will be readily understood by one of ordinary skill in the art, a urine sample must be “obtained” prior to assaying the sample. Thus, the term “assaying” implies that the sample has been obtained. The terms “obtained” or “obtaining” as used herein encompass the act of receiving an extracted or isolated urine sample. For example, a testing facility can “obtain” a urine sample via delivery prior to assaying the sample. In some such cases, the urine sample was “extracted” or “isolated” from a renal transplant recipient by another party prior to delivery (transfer, etc.), and then “obtained” by the testing facility upon arrival of the sample. Thus, a testing facility can obtain the sample and then assay the sample, thereby producing data related to the sample.

[0199] The terms “obtained” or “obtaining” as used herein can also include the physical extraction of a urine sample from a renal transplant recipient. Accordingly, a urine sample can be extracted (and thus “obtained”) by the same person or same entity that subsequently assays the sample. When a urine sample is “extracted” by a first party or entity and then transferred (e.g., delivered, mailed, etc.) to a second party, the sample was “obtained” by the first party (and also “extracted” by the first party), and then subsequently “obtained” (but not “extracted”) by the second party. Accordingly, in some embodiments, the step of obtaining does not comprise the step of extracting a urine sample.

[0200] It will be understood by one of ordinary skill in the art that in some cases, it is convenient to wait until multiple samples have been obtained prior to assaying the samples. Accordingly, in some cases a urine sample is stored until all appropriate samples have been obtained. One of ordinary skill in the art will understand how to appropriately store urine samples and any convenient method of storage may be used (e.g., refrigeration) that is appropriate for the particular urine sample. In some embodiments, a baseline urine sample (i.e., before starting immunosuppressive therapy) is assayed prior to obtaining another urine sample, obtained at a later time point during immunosuppressive therapy. In some cases, a baseline urine sample and a urine sample, obtained at a later time point during immunosuppressive therapy, are assayed in parallel. In some cases, multiple urine samples, taken at different time points during immunosuppressive therapy, are assayed in parallel. In some cases, urine samples are processed immediately or as soon as possible after they are obtained.

[0201] In some embodiments, the concentration (i.e., “level”) of a metabolite (which will be referenced herein as a biomarker) in a urine sample is measured (i.e., “determined”). By “level” (or “concentration”) it is meant the level of the biomarker (e.g., the absolute and / or normalized value determined for the level of a biomarker in a urine sample).

[0202] The terms “determining”, “measuring”, “evaluating”, “assessing,” “assaying,” and “analyzing” are used interchangeably herein to refer to any form of measurement, and include determining if an element is present or not. These terms include both quantitative and / or qualitative determinations. Assaying may be relative or absolute. For example, “assaying” can be determining whether the level is less than or “greater than or equal to” a particular threshold, (the threshold can be pre-determined or can be determined by assaying a control sample). On the other hand, “assaying to determine the “level” can mean determining a quantitative value (using any convenient metric) that represents the level (i.e., concentration or amount of a metabolite) of a particular biomarker. The level can be expressed in arbitrary units associated with a particular assay (e.g., NMR or LC-MS signal intensity), or can be expressed as an absolute value with defined units (e.g., number of metabolite molecules, concentration of metabolite, etc.). Additionally, the level of a biomarker can be compared to the level of one or more additional biomarkers to derive a normalized value that represents a normalized level. The specific metric (or units) chosen is not crucial as long as the same units are used (or conversion to the same units is performed) when evaluating multiple urine samples from the same renal transplant recipient (e.g., urine samples taken at different points in time from the same renal transplant recipient). This is because the units cancel when calculating a fold-change (i.e., determining a ratio) in the level from one urine sample to the next (e.g., urine samples taken at different points in time from the same renal transplant recipient).

[0203] In some instances, the concentration of one or more metabolite biomarkers may be measured, and a biomarker concentration is compared to the level of one or more additional metabolites to provide a normalized value for the biomarker concentration. Any convenient protocol for evaluating metabolite levels may be employed wherein the level of one or more metabolites in the assayed urine sample is determined.

[0204] Various methods for measuring the levels of metabolite biomarkers in a urine sample may be employed. Representative exemplary methods include but are not limited to nuclear magnetic resonance (NMR), mass spectrometry (MS), liquid chromatography-mass spectrometry (LC-MS), tandem mass spectrometry (MS / MS), gas chromatography-mass spectrometry (GC-MS), vibrational spectroscopy (e.g., mid-infrared (IR) and Raman), spectrophotometric assays, enzymatic or biochemical assays, and liquid chromatography.Additional terms.

[0205] The term "about," particularly in reference to a given quantity, is meant to encompass deviations of plus or minus five percent.

[0206] The terms “recipient”, “individual”, “subject”, and “patient” are used interchangeably herein and refer to any mammalian subject who has received a renal transplant, particularly humans. Mammalian subjects include human and non-human mammals such as non-human primates, including chimpanzees and other apes and monkey species; laboratory animals such as mice, rats, rabbits, hamsters, guinea pigs, and chinchillas; domestic animals such as dogs and cats; and farm animals such as sheep, goats, pigs, horses, and cows.

[0207] “Isolated” refers to an entity of interest that is in an environment different from that in which it may naturally occur. “Isolated” is meant to include entities that are within samples that are substantially enriched for the entity of interest and / or in which the entity of interest is partially or substantially purified.

[0208] "Substantially purified" generally refers to isolation of a component such as a substance (compound, metabolite) such that the substance comprises the majority percent of the sample in which it resides. Typically in a sample, a substantially purified component comprises at least 50%, preferably at least 80%-85%, more preferably at least 90-95% of the sample.

[0209] “Providing an analysis” is used herein to refer to the delivery of an oral or written analysis (i.e., a document, a report, etc.). A written analysis can be a printed or electronic document. A suitable analysis (e.g., an oral or written report) provides any or all of the following information: identifying information of the renal transplant recipient (name, age, etc.), a description of what type of urine sample(s) was used and / or how it was used, the technique used to assay the sample (e.g., NMR and / or LC-MS), the results of the assay (e.g., the level of the biomarker as measured, and / or the fold-change of a biomarker level over time), an assessment as to whether the renal transplant recipient is determined to have over-immunosuppression, under-immunosuppression, graft dysfunction, subclinical graft rejection, graft rejection, or an infection, a recommendation to continue or alter current immunosuppressive therapy, a recommended strategy for additional therapy, etc. The report can be in any format including, but not limited to printed information on a suitable medium or substrate (e.g., paper); or electronic format. If in electronic format, the report can be in any computer readable medium, e.g., diskette, compact disk (CD), flash drive, and the like, on which the information has been recorded. In addition, the report may be present as a website address which may be used via the internet to access the information at a remote site.

[0210] The terms "treatment", "treating", "treat" and the like are used herein to generally refer to obtaining a desired pharmacologic and / or physiologic effect. The effect can be prophylactic in terms of completely or partially preventing a disease or symptom(s) thereof and / or may be therapeutic in terms of a partial or complete stabilization or cure for a disease and / or adverse effect attributable to the disease. The term “treatment" encompasses any treatment of a disease in a mammal, particularlya human, and includes: (a) preventing the disease and / or symptom(s) from occurring in a subject who may be predisposed to the disease or symptom but has not yet been diagnosed as having it; (b) inhibiting the disease and / or symptom(s), i.e., arresting their development; or (c) relieving the disease symptom(s), i.e., causing regression of the disease and / or symptom(s). Those in need of treatment include those already inflicted (e.g., those with over-immunosuppression, underimmunosuppression, graft dysfunction, subclinical graft rejection, graft rejection, or an infection) as well as those in which prevention is desired (e.g., those treated with immunosuppressive therapy to prevent transplant rejection who are at risk of over-immunosuppression, under-immunosuppression, graft dysfunction, subclinical graft rejection, graft rejection, or an infection).

[0211] A therapeutic treatment is one in which the subject is inflicted prior to administration and a prophylactic treatment is one in which the subject is not inflicted prior to administration. In some embodiments, the subject has an increased likelihood of becoming inflicted or is suspected of being inflicted prior to treatment. In some embodiments, the subject is suspected of having an increased likelihood of becoming inflicted.Biomarkers for Evaluating Whether a Renal Transplant Recipient is Over-Immunosuppressed or Under-Immunosuppressed

[0212] The inventors have discovered urine-based metabolite signatures that distinguish overimmunosuppressed renal transplant recipients and under-immunosuppressed renal transplant recipients from those with a stable graft with high accuracy. In some embodiments, the subject methods utilize nuclear magnetic resonance (NMR) spectroscopy and / or liquid chromatographymass spectrometry (LC-MS) to detect metabolite features that differentiate over-immunosuppressed patients from under-immunosuppressed patients and patients who have stable grafts.

[0213] Metabolite biomarkers include, but are not limited to, lysine, caffeine, benzoyl formic acid, cytidine, glucuronic acid, glucaric acid, lactose, glucose-6-phosphate, 1 -methylhistamine, histidine, trigonelline, trimethylamine N-oxide (TMAO), hippuric acid, choline, phosphoethanolamine, phosphocholine, acetylglycine, phenyllactic acid, mycophenolic acid glucuronide (MPAG), creatine, and hydroxyphenylacetic acid. The levels of these metabolite biomarkers in urine samples differ for renal transplant recipients who are over-immunosuppressed, under-immunosuppressed, or stable (e.g., an individual known to not have over-immunosuppression, under-immunosuppression, graft dysfunction, subclinical graft rejection, graft rejection, or an infection). Accordingly, these metabolite biomarkers are useful for monitoring a renal transplant recipient to avoid over-immunosuppression, which can lead to infections that cause graft dysfunction or loss or even death of the patient as wellas monitoring a renal transplant recipient to avoid under-immunosuppression, which can lead to graft dysfunction, subclinical graft rejection, or graft rejection.

[0214] A urine sample comprising metabolite biomarkers is obtained from a renal transplant recipient. A "control" sample, as used herein, refers to a urine sample from a renal transplant recipient that is not over-immunosuppressed or under-immunosuppressed. That is, a control sample is obtained from a renal transplant recipient who is stable (e.g., an individual known to not have overimmunosuppression, under-immunosuppression, graft dysfunction, subclinical graft rejection, graft rejection, or an infection). A urine sample can be obtained from a subject by conventional techniques. For example, urine samples can be obtained by having the subject urinate into a collection container or by collection of urine using a urinary catheter according to methods well known in the art.

[0215] Urine samples may be manipulated in any way after their procurement, such as by treatment with reagents, washed, or enriched for particular types of molecules, e.g., metabolite biomarkers. In some embodiments, metabolites are extracted using precipitation with an organic solvent. For example, methanol, acetonitrile / methanol, methanol / chloroform, or chloroform, among others, may be used to precipitate proteins and other macromolecules, which can be removed from solution by centrifugation, wherein the levels of the metabolites remaining in the supernatant are measured. In some embodiments, filtration is used to separate and isolate metabolites. The solution containing metabolites may be partially evaporated under reduced pressure and / or lyophilized. In some cases, a solution enriched with metabolites is completely dried, and the metabolites are resuspended in a different solvent prior to measuring the levels of the metabolites.

[0216] In some embodiments, urine samples are obtained from a renal transplant recipient before, during, or after treatment with immunosuppressive therapy. In some embodiments, urine samples are obtained from the subject periodically throughout immunosuppressive therapy to determine whether the renal transplant recipient is becoming over-immunosuppressed or underimmunosuppressed. In some embodiments, urine samples are obtained at least twice a month, at least once a month, at least every 2 months, at least every 3 months, at least every 4 months, at least every 5 months, at least every 6 months, or at least once a year.

[0217] In some embodiments, NMR is used for detection of biomarkers. Identification of metabolite biomarkers may be performed manually or automated or semi-automated by computer-aided spectral deconvolution and spectral matching to a library of reference NMR spectra. In some embodiments, 1 D NMR (e.g.,1H,13C,31P, or15N NMR) is used to detect and quantitate biomarkers. However, 1 D NMR may suffer from poor resolution. Overlapping NMR signals can be resolved by increasing the number of dimensions (e.g., switching from 1 D to 2D or 3D spectra) to detect metabolite biomarkers. In some embodiments, 2D NMR techniques such as 2D1H-13C heteronuclearsingle quantum coherence (HSQC) or 2D1H-1H-total correlation spectroscopy (TOCSY) are used to detect metabolite biomarkers. NMR pulse sequences such as heteronuclear single quantum coherence (HSQC), HSQCo, Q-HSQC, QQ-HSQC, and quantitative perfected and pure shifted HSQC or QUIPU HSQC can be used to reduce peak variability arising from differences in coupling constants and other parameters. In some embodiments, non-uniformed sampling (NUS) multidimensional NMR is used to enhance resolution and sensitivity of detection of metabolite biomarkers. For a description of NMR methods for detecting metabolites, see, e.g., (Hoyt & O’Day (2021) Magn Reson Chem 59(3):257-263, Du et al. (2023) Anal Chem. 95(6):3195-3203, Weitzel et al. (2020) Metabolites 10(1 1):449., Vignoli et al. (2019) Angew Chem Int Ed Engl 58(4):968-994, Wishart et al. (2022) Metabolites 12(8):678, Moco et al. (2022) Front Mol Biosci 9:882487, Zhang et al. (2013) Magn Reson Chem. 51 (9):549-56; herein incorporated by reference in their entireties.

[0218] In certain embodiments, one or more differential NMR metabolite features are used to classify the renal transplant recipient as over-immunosuppressed or stable. By “differential NMR metabolite feature” is meant a spectral feature that is differentially present or at a different intensity in NMR spectra of metabolites from urine samples of renal transplant recipients who are overimmunosuppressed compared to stable renal transplant recipients who are not overimmunosuppressed. See, e.g., FIGS. 8A-8C for exemplary NMR metabolite features that can be used to distinguish between a renal transplant recipient who is over-immunosuppressed and a stable renal transplant recipient who is not over-immunosuppressed.

[0219] In certain embodiments, one or more differential NMR metabolite features, used to classify the renal transplant recipient as over-immunosuppressed or stable, are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.94 13C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.

[0220] In certain embodiments, the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1 103 having a chemical shift at 3.961H ppm and a chemical shift at 72.85 13C ppm.

[0221] In certain embodiments, the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.

[0222] In some embodiments, mass spectrometry is used for detection of biomarkers. Mass spectrometry detection of molecules typically involves formation of gas phase ions through electron impact ionization (El), electrospray ionization (ESI), heated electrospray ionization (HESI), or matrix- assisted laser desorption / ionization (MALDI). In some cases, metabolites in urine samples are separated prior to ionization using gas chromatography (GC-MS) or liquid chromatography (LC-MS). For a description of mass spectrometry techniques for identifying metabolites, see, e.g., Alseekh et al. (2021 ) Nat Methods 18(7):747-756, Zhou et al. (2012) Mol Biosyst. 8(2):470-81 , Chen et al. (2022) Mass Spectrom Rev. 29:e21785, Heiles (2021 ) Anal Bioanal Chem. 413(24):5927-5948, Zeki et al. (2020) J Pharm Biomed Anal. 190:113509, Wright (2011 ) Xenobiotica 41 (8):670-686; herein incorporated by reference in their entireties.

[0223] In certain embodiments, one or more differential LC-MS metabolite features are used to classify the renal transplant recipient as over-immunosuppressed or stable. By “differential LC-MS metabolite feature” is meant a spectral feature that is differentially present or at a different intensity in LC-MS spectra of metabolites from urine samples of renal transplant recipients who are overimmunosuppressed compared to stable renal transplant recipients who are not overimmunosuppressed. See, e.g., FIG. 16 for exemplary LC-MS metabolite features that can be used to distinguish between a renal transplant recipient who is over-immunosuppressed and a stable renal transplant recipient who is not over-immunosuppressed.

[0224] In certain embodiments, one or more differential LC-MS metabolite features of caffeine, benzoyl formic acid, and cytidine are used to classify the renal transplant recipient as overimmunosuppressed or stable using the machine learning model. In some embodiments, the differential LC-MS metabolite features for caffeine comprise a feature with a mass-to-charge ratio (m / z) of 195.00 for a caffeine ([M+H]+ ion and a feature with a m / z of 137.90 and a feature with a m / z of 110.00 from selected reaction monitoring. In some embodiments, the differential LC-MS metabolite features for benzoyl formic acid comprise a feature with a m / z of 149.09 for a benzoyl formic acid [M-H]- ion and a feature with a m / z of 104.91 and a feature with a m / z of 77.07 from selected reaction monitoring. In some embodiments, the differential LC-MS metabolite features forcytidine comprise a feature with a m / z of 244.09 for a cytidine [M+H]+ ion and a feature with a m / z of 111 .90 from selected reaction monitoring.

[0225] In certain embodiments, one or more differential NMR metabolite features are used to classify the renal transplant recipient as under-immunosuppressed or stable. By “differential NMR metabolite feature” is meant a spectral feature that is differentially present or at a different intensity in NMR spectra of metabolites from urine samples of renal transplant recipients who are underimmunosuppressed compared to stable renal transplant recipients who are not underimmunosuppressed. See, e.g., Table 4 for exemplary NMR metabolite features that can be used to distinguish between a renal transplant recipient who is under-immunosuppressed and a stable renal transplant recipient who is not under-immunosuppressed.

[0226] In certain embodiments, the one or more differential NMR metabolite features, used to classify the renal transplant recipient as under-immunosuppressed or stable, are selected from X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X1223 having a chemical shift at 7.941H ppm and a chemical shift at 140.3013C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51 .0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.

[0227] In certain embodiments, the differential NMR metabolite features, used to classify the renal transplant recipient as under-immunosuppressed or stable, comprise or consist of X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51 .0413C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.55 13C ppm.

[0228] To allow comparisons to be made between biomarker levels measured in different urine samples or by different techniques, the levels of metabolite biomarkers can be compared to the levels of one or more additional metabolites to derive normalized values for the levels of the biomarkers. A metabolite selected for use in normalization (i.e., “normalization metabolite”) should have the same level in a urine sample whether the renal transplant recipient is overimmunosuppressed, under-immunosuppressed, or stable (i.e., not differentially expressed).

[0229] In certain embodiments, the method comprises using one or more normalization NMR metabolite features in combination with the differential NMR metabolite features to classify the renaltransplant recipient as over-immunosuppressed, under-immunosuppressed, or stable using a machine learning model. In some embodiments, the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.94 13C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.82 ’H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.56 ’H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm. In some embodiments, the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.06 ’H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1 .561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm. In certain embodiments, the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.

[0230] In certain embodiments, the method further comprises using one or more normalization LC- MS metabolite features in combination with the differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed, under-immunosuppressed, or stable using a machine learning model. In certain embodiments, the normalization LC-MS metabolite features comprise lysine, hippuric acid, mannitol, trimethylamine N-oxide, and glutamine LC-MS features.

[0231] In certain embodiments, the one or more differential NMR metabolite features are used in combination with one or more differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed, under-immunosuppressed, or stable using the machine learning model, wherein the one or more differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppmand a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm; wherein the one or more differential NMR metabolite features used to classify the renal transplant recipient as underimmunosuppressed or stable are selected from X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X1223 having a chemical shift at 7.941H ppm and a chemical shift at 140.3013C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51 .0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm; and wherein the one or more differential LC-MS metabolite features used to classify the renal transplant recipient as over-immunosuppressed are selected from a feature with a mass-to-charge ratio (m / z) of 195.00 for a caffeine ([M+H]+ ion and a feature with a m / z of 137.90 and a feature with a m / z of 1 10.00 from selected reaction monitoring, a feature with a m / z of 149.09 for a benzoyl formic acid [M-H]- ion and a feature with a m / z of 104.91 and a feature with a m / z of 77.07 from selected reaction monitoring, a feature with a m / z of 244.09 for a cytidine [M+H]+ ion and a feature with a m / z of 1 11 .90 from selected reaction monitoring.

[0232] In certain embodiments, the method further comprises using one or more normalization NMR metabolite features and normalization LC-MS metabolite features (as described above) in combination with the differential NMR metabolite features and differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed, under-immunosuppressed, or stable using the machine learning model.

[0233] Metabolites levels change over time after a kidney transplant. Therefore, the time posttransplant is a variable that may be considered in selecting biomarkers and building a classifier to distinguish between samples indicating a graft is at risk of injury or is stable. In some embodiments, metabolite biomarkers and clinical information are used in combination to determine if a kidney graft is stable or unstable. Such clinical information may include, but is not limited to, the gender of the renal transplant recipient, the age of the renal transplant recipient at the time of the transplant of the kidney graft, the plasma level of BK polyomavirus (BKV) in the renal transplant recipient (e.g., log BKV PCR), level of serum creatinine in the renal transplant recipient, and estimated glomerular filtration rate (eGFR) of the renal transplant recipient

[0234] In certain embodiments, a method of monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy is provided, the method comprising: (a) obtaining a urine sample from the renal transplant recipient; and (b) measuring levels of one or more metabolites in the urine sample using nuclear magnetic resonance (NMR), wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, intensities of one or more differential NMR metabolite features of one or more metabolites in the urine sample are used to classify the kidney graft as stable or unstable at risk of injury using a first machine learning model, wherein the one or more differential NMR metabolite features are selected from X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.44 ’H ppm and a chemical shift at 51.0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm; and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, clinical information comprising gender of the renal transplant recipient, age of the renal transplant recipient at time of transplant of the kidney graft, plasma level of BK polyomavirus (BKV) in the renal transplant recipient, level of serum creatinine in the renal transplant recipient, and estimated glomerular filtration rate (eGFR) of the renal transplant recipient in combination with intensities of one or more differential NMR metabolite features of one or more metabolites in the urine sample are used to classify the kidney graft as stable or unstable at risk of injury using a second machine learning model, wherein the one or more differential NMR metabolite features are selected from an NMR metabolite feature having a chemical shift at 4.261H ppm and a chemical shift at 68.9913C ppm, an NMR metabolite feature having a chemical shift at 3.87 ’H ppm and a chemical shift at 58.813C ppm, an NMR metabolite feature having a chemical shift at 7.141H ppm and a chemical shift at 1 19.9313C ppm, an NMR metabolite feature having a chemical shift at 3.221H ppm and a chemical shift at 30.1213C ppm, an NMR metabolite feature having a chemical shift at 8.181H ppm and a chemical shift at 137.9213C ppm, and one or more NMR metabolite features for one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3- dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine.

[0235] In certain embodiments, wherein the kidney graft is classified as unstable at risk of injury, the method further comprises determining whether the renal transplant recipient is underimmunosuppressed or over-immunosuppressed, wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, clinical information comprising the age of the renal transplant recipient at the time of transplant of the kidney graft and the plasma level of BKV in the renal transplant recipient in combination with intensities of differential NMR metabolite features comprising X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm and X2473 having a chemical shift at 3.811H ppm and a chemical shift at 63.9313C ppm are used to classify the renal transplant recipient as under-immunosuppressed or over-immunosuppressed using a third machine learning model; and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, clinical information comprising the gender of the renal transplant recipient, the age of the renal transplant recipient at time of transplant of the kidney graft, the plasma level of BKV in the renal transplant recipient, the level of serum creatinine in the renal transplant recipient, and the eGFR of the renal transplant recipient in combination with intensities of the one or more differential NMR metabolite features selected from the NMR metabolite feature having the chemical shift at 4.261H ppm and the chemical shift at 68.9913C ppm, the NMR metabolite feature having the chemical shift at 3.871H ppm and the chemical shift at 58.813C ppm, the NMR metabolite feature having the chemical shift at 7.141H ppm and the chemical shift at 1 19.9313C ppm, the NMR metabolite feature having the chemical shift at 3.221H ppm and the chemical shift at 30.1213C ppm, the NMR metabolite feature having the chemical shift at 8.181H ppm and the chemical shift at 137.9213C ppm, and the one or more NMR metabolite features for the one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine are used to classify the renal transplant recipient as under-immunosuppressed or overimmunosuppressed using a fourth machine learning model.

[0236] In some embodiments, the methods described herein are used repeatedly for monitoring a renal transplant recipient undergoing treatment with immunosuppressive therapy to determine if the renal transplant recipient is becoming over-immunosuppressed or under-immunosuppressed over time. In some embodiments levels of metabolite biomarkers are measured in urine samples at set intervals. In some embodiments, the levels of metabolite biomarkers are measured at least twice a month, at least once a month, at least every 2 months, at least every 3 months, at least every 4months, at least every 5 months, at least every 6 months, or at least once a year and analyzed, as described herein, to determine if the renal transplant recipient is becoming over-immunosuppressed or under-immunosuppressed. In some embodiments, levels of metabolite biomarkers are measured in urine samples if the subject shows symptoms of an infection, graft dysfunction, subclinical graft rejection, or graft rejection.

[0237] The determination that a renal transplant recipient is over-immunosuppressed or underimmunosuppressed based on analysis of levels of metabolite biomarkers in urine is an active clinical application of the correlation between levels of the metabolite biomarkers and overimmunosuppression or under-immunosuppression. For example, “determining” requires the active step of reviewing the data, which is produced during the active assaying step(s), and resolving whether an individual does or does not have over-immunosuppression or underimmunosuppression. Additionally, in some cases, a decision is made to proceed with the current immunosuppressive therapy, or instead to alter the treatment. In some cases, the subject methods include the step of continuing therapy or altering immunosuppressive therapy.

[0238] The term “continue treatment” (i.e., continue therapy) is used herein to mean that the current course of treatment (e.g., continued administration of immunosuppressive therapy with the current regimen) is to continue. If the current course of treatment is causing over-immunosuppression, the treatment may be altered. “Altering therapy” is used herein to mean “changing the immunosuppressive therapy” (e.g., changing the type of immunosuppressive agent(s) administered, changing the particular dose and / or frequency of administration of an immunosuppressive agent, e.g., decreasing the dose and / or frequency if over-immunosuppression is detected based on the levels of the metabolite biomarkers or increasing the dose and / or frequency if underimmunosuppression is detected based on the levels of the metabolite biomarkers). In some cases, immunosuppressive therapy can be altered until the individual is deemed to be stable (e.g., no overimmunosuppression, no under-immunosuppression, no infection, no graft dysfunction, no subclinical graft rejection or graft rejection). In some embodiments, altering therapy means changing which type of treatment is administered, or discontinuing a particular treatment altogether, etc.

[0239] As a non-limiting illustrative example, a renal transplant recipient may be initially treated with immunosuppressive therapy by administering a corticosteroid (e.g., prednisone, prednisolone, or hydrocortisone), a calcineurin inhibitor (e.g., cyclosporine or tacrolimus), and an antiproliferative agent (e.g., mycophenolate mofetil, mycophenolate sodium, or azathioprine). Then to “continue treatment” would be to continue with this type of treatment. If the current course of treatment is causing over-immunosuppression, the treatment may be altered, e.g., by decreasing the dose or frequency of administration of one or more immunosuppressive agents or eliminating treatment withone or more immunosuppressive agents and adding treatment with another immunosuppressive agent of the same type or a different type. For example, an mTOR inhibitor (e.g., sirolimus) may be added to the treatment while reducing or eliminating treatment with a calcineurin inhibitor or a steroid. If during the course of treatment, the renal transplant recipient becomes under-immunosuppressed, the treatment may be altered, e.g., by increasing the dose or frequency of administration of one or more immunosuppressive agents or adding treatment with another immunosuppressive agent of the same type or a different type.

[0240] In other words, the levels of one or more metabolite biomarkers may be monitored in order to determine when to continue immunosuppressive therapy with the current regimen and / or when to alter immunosuppressive therapy. As such, a post-treatment urine sample can be isolated after any of the administrations of immunosuppressive agents, and the urine sample can be assayed to determine the levels of metabolites biomarkers in order to determine if the subject is becoming overimmunosuppressed or under-immunosuppressed.

[0241] In some embodiments, the subject methods include providing an analysis of whether a renal transplant recipient is over-immunosuppressed or under-immunosuppressed. A suitable analysis (e.g., an oral or written report) provides any or all of the following information: identifying information of the renal transplant recipient (name, age, etc.), a description of what type of urine sample(s) was used and / or how it was used, the technique used to assay the sample (e.g., NMR and / or LC-MS), the results of the assay (e.g., the level of the biomarker as measured, and / or the fold-change of a biomarker level over time), an assessment as to whether the renal transplant recipient is determined to have over-immunosuppression, under-immunosuppression, graft dysfunction, subclinical graft rejection, graft rejection, or an infection, a recommendation to continue or alter current immunosuppressive therapy, a recommended strategy for additional therapy, etc. As described above, an analysis can be an oral or written report (e.g., written or electronic document). The analysis can be provided to the renal transplant recipient, to the subject’s physician, to a testing facility, etc. The analysis can also be accessible as a website address via the internet. In some such cases, the analysis can be accessible by multiple different entities (e.g., the subject, the subject’s physician, a testing facility, etc.).Data Analysis

[0242] Analyzing the levels of a plurality of metabolite biomarkers may comprise the use of an algorithm or classifier. In some embodiments, a machine learning algorithm is used to classify a renal transplant recipient undergoing treatment with immunosuppressive therapy as overimmunosuppressed, under-immunosuppressed, or stable (e.g., not over-immunosuppressed orunder-immunosuppressed, no infection, graft dysfunction, subclinical graft rejection, or graft rejection) based on the levels of metabolite biomarkers. The machine learning algorithm may comprise a supervised learning algorithm. Examples of supervised learning algorithms may include Average One-Dependence Estimators (AODE), Artificial neural network (e.g., Backpropagation), Bayesian statistics (e.g., Naive Bayes classifier, Bayesian network, Bayesian knowledge base), Case-based reasoning, Decision trees, Inductive logic programming, Gaussian process regression, Group method of data handling (GMDH), Learning Automata, Learning Vector Quantization, Minimum message length (decision trees, decision graphs, etc.), Lazy learning, Instance-based learning Nearest Neighbor Algorithm, Analogical modeling, Probably approximately correct learning (PAC) learning, Ripple down rules, a knowledge acquisition methodology, Symbolic machine learning algorithms, Subsymbolic machine learning algorithms, Support vector machines, Random Forests, Ensembles of classifiers, Bootstrap aggregating (bagging), and Boosting. Supervised learning may comprise ordinal classification such as regression analysis and Information fuzzy networks (IFN). Alternatively, supervised learning methods may comprise statistical classification, such as AODE, Linear classifiers (e.g., Fisher's linear discriminant, Logistic regression, Naive Bayes classifier, Perceptron, and Support vector machine), quadratic classifiers, k-nearest neighbor, Boosting, Decision trees (e.g., C4.5, Random forests), Bayesian networks, and Hidden Markov models.

[0243] The machine learning algorithms may also comprise an unsupervised learning algorithm. Examples of unsupervised learning algorithms may include artificial neural network, Data clustering, Expectation-maximization algorithm, Self-organizing map, Radial basis function network, Vector Quantization, Generative topographic map, Information bottleneck method, and IBSEAD. Unsupervised learning may also comprise association rule learning algorithms such as Apriori algorithm, Eclat algorithm and FP-growth algorithm. Hierarchical clustering, such as Single-linkage clustering and Conceptual clustering, may also be used. Alternatively, unsupervised learning may comprise partitional clustering such as K-means algorithm and Fuzzy clustering.

[0244] In some instances, the machine learning algorithms comprise a reinforcement learning algorithm. Examples of reinforcement learning algorithms include, but are not limited to, temporal difference learning, Q-learning and Learning Automata. Alternatively, the machine learning algorithm may comprise Data Pre-processing.

[0245] In some embodiments, the machine learning algorithm uses artificial neural networks. In some embodiments, the machine learning algorithm uses a deep learning algorithm, which may include the use of convolutional neural networks, deep neural networks, recurrent neural networks, efficient neural networks, deep residual neural networks, long short-term memory networks, deepbelief networks, multilayer perceptrons, or deep reinforcement learning, and the like. See, e.g., Pedrycz et al. Deep Learning: Algorithms and Applications (Studies in Computational Intelligence Book 865, Springer, 2019), Goodfellow et al. Deep Learning (Adaptive Computation and Machine Learning series, The MIT Press, 2016), and Various Deep Learning Algorithms in Computational Intelligence (edited by Oscar Humberto Montiel Ross, Mdpi AG, 2023); herein incorporated by reference in their entireties.

[0246] In certain embodiments, the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA), or a random forest (RF) machine learning algorithmSystem and Computer Implemented Methods for Evaluating Whether a Renal Transplant Recipient is Over-Immunosuppressed or Under-lmmunosuppressed

[0247] The present disclosure also provides systems and computer implemented methods which find use in practicing the subject methods. In some embodiments, a computer implemented method is used for evaluating whether a renal transplant recipient is over-immunosuppressed. The processor can be programmed to perform steps of a computer implemented method comprising: a) receiving NMR and / or LC-MS spectra of one or more metabolites from a urine sample obtained from the renal transplant recipient; b) measuring levels of the one or more metabolites using the NMR and / or LC- MS spectra; c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed; and d) displaying information regarding whether the renal transplant recipient is over-immunosuppressed. In certain embodiments, the computer implemented method further comprises storing the information regarding whether the renal transplant recipient is over-immunosuppressed in a database.

[0248] In certain embodiments, the NMR spectra are multi-dimensional NMR spectra. In certain embodiments, the NMR spectra comprise one-dimensional (1 D)1H NMR spectra, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectra, or a combination thereof. In certain embodiments, the NMR spectra are obtained using non-uniformed sampling (NUS).

[0249] In certain embodiments, one or more differential NMR metabolite features in the NMR spectra are used to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model, wherein the one or more differential NMR metabolite features are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppmand a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.

[0250] In certain embodiments, the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.85 13C ppm.

[0251] In certain embodiments, the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.

[0252] In certain embodiments, the computer implemented method further comprises using one or more normalization NMR metabolite features in combination with the differential NMR metabolite features in the NMR spectra to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

[0253] In certain embodiments, one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71.9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

[0254] In certain embodiments, the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm anda chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.72 13C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.78 ’H ppm and a chemical shift at 57.0813C ppm.

[0255] In certain embodiments, the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.

[0256] In certain embodiments, the chemical shifts are calibrated against an internal deuterated sodium 2,2-dimethyl-2-silapentane-5-sulfonate standard.

[0257] In certain embodiments, one or more differential LC-MS metabolite features of caffeine, benzoyl formic acid, and cytidine in the LC-MS spectra are used to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

[0258] In certain embodiments, the differential LC-MS metabolite features for caffeine comprise a feature with a mass-to-charge ratio (m / z) of 195.00 for a caffeine ([M+H]+ ion and a feature with a m / z of 137.90 and a feature with a m / z of 110.00 from selected reaction monitoring.

[0259] In certain embodiments, the differential LC-MS metabolite features for benzoyl formic acid comprise a feature with a m / z of 149.09 for a benzoyl formic acid [M-H]- ion and a feature with a m / z of 104.91 and a feature with a m / z of 77.07 from selected reaction monitoring.

[0260] In certain embodiments, the differential LC-MS metabolite features for cytidine comprise a feature with a m / z of 244.09 for a cytidine [M+H]+ ion and a feature with a m / z of 111 .90 from selected reaction monitoring.

[0261] In some embodiments, a computer implemented method is used for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is overimmunosuppressed, under-immunosuppressed, or stable, the computer performing steps comprising: (a) receiving nuclear magnetic resonance (NMR) and / or mass spectrometry spectra of one or more metabolites from a urine sample obtained from the renal transplant recipient; (b) measuring levels of the one or more metabolites using the NMR spectra and / or mass spectrometry spectra; (c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed, underimmunosuppressed, or stable; and (d) displaying information regarding whether the renal transplant recipient is over-immunosuppressed, under-immunosuppressed, or stable.

[0262] In certain embodiments, one or more differential NMR metabolite features are used to classify the renal transplant recipient as over-immunosuppressed, under-immunosuppressed, or stable using the machine learning model, wherein the one or more differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1 103 having a chemical shift at 3.961H ppm and a chemical shift at 72.85 13C ppm; and wherein the one or more differential NMR metabolite features used to classify the renal transplant recipient as under-immunosuppressed or stable are selected from X65 having a chemical shift at 4.06 ’H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X1223 having a chemical shift at 7.941H ppm and a chemical shift at 140.3013C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.

[0263] In certain embodiments, the differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.

[0264] In certain embodiments, the differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.

[0265] In certain embodiments, the differential NMR metabolite features used to classify the renal transplant recipient as under-immunosuppressed or stable comprise or consist of X65 having achemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.67 ’H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.17 ’H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51 .0413C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.55 13C ppm.

[0266] In certain embodiments, the computer implemented method further comprises using one or more normalization LC-MS metabolite features in combination with the differential LC-MS metabolite features in the LC-MS spectra to classify the renal transplant recipient as over-immunosuppressed, under-immunosuppressed, or stable using the machine learning model.

[0267] In certain embodiments, the normalization LC-MS metabolite features comprise lysine, hippuric acid, mannitol, trimethylamine N-oxide, and glutamine LC-MS features.

[0268] In certain embodiments, the one or more differential NMR metabolite features are used in combination with one or more differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed, under-immunosuppressed, or stable using the machine learning model, wherein the one or more differential NMR metabolite features are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.85 13C ppm; and wherein the one or more differential LC-MS metabolite features are selected from a feature with a mass-to-charge ratio (m / z) of 195.00 for a caffeine ([M+H]+ ion and a feature with a m / z of 137.90 and a feature with a m / z of 110.00 from selected reaction monitoring, a feature with a m / z of 149.09 for a benzoyl formic acid [M-H]- ion and a feature with a m / z of 104.91 and a feature with a m / z of 77.07 from selected reaction monitoring, a feature with a m / z of 244.09 for a cytidine [M+H]+ ion and a feature with a m / z of 111 .90 from selected reaction monitoring.

[0269] In certain embodiments, the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.85 13C ppm.

[0270] In certain embodiments, the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.

[0271] In some embodiments, a computer implemented method is used for evaluating whether a renal transplant recipient is under-immunosuppressed. The processor can be programmed to perform steps of a computer implemented method comprising: a) receiving NMR and / or LC-MS spectra of one or more metabolites from a urine sample obtained from the renal transplant recipient; b) measuring levels of the one or more metabolites using the NMR and / or LC-MS spectra; c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is under-immunosuppressed; and d) displaying information regarding whether the renal transplant recipient is under-immunosuppressed. In certain embodiments, the computer implemented method further comprises storing the information regarding whether the renal transplant recipient is under-immunosuppressed in a database.

[0272] In certain embodiments, the computer implemented method further comprises storing the information regarding whether the renal transplant recipient is under-immunosuppressed or stable in a database.

[0273] In certain embodiments, the computer implemented method further comprises using one or more normalization NMR metabolite features and normalization LC-MS metabolite features in combination with the differential NMR metabolite features and differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed, under-immunosuppressed, or stable using the machine learning model.

[0274] In certain embodiments, the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71.9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

[0275] In certain embodiments, the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.72 13C ppm, X59 having a chemical shift at 5.24 ’H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.83 ’H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.78 1H ppm and a chemical shift at 57.0813C ppm.

[0276] In certain embodiments, the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.

[0277] In certain embodiments, the normalization LC-MS metabolite features comprise lysine, hippuric acid, mannitol, trimethylamine N-oxide, and glutamine LC-MS features.

[0278] In certain embodiments, the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA) or a random forest (RF) machine learning algorithm.

[0279] In certain embodiments, a computer implemented method for monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy is provided, the computer performing steps comprising: (a) receiving nuclear magnetic resonance (NMR) spectra of one or more metabolites from a urine sample obtained from the renal transplant recipient; (b) measuring intensities of one or more differential NMR metabolite features in the NMR spectra, wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, the intensities of the one or more differential NMR metabolite features of the one or more metabolites in the urine sample are used to classify the kidney graft as stable or unstable at risk of injury using a first machine learning model, wherein the one or more differential NMR metabolite features are selected from X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.67 ’H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.17 ’H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51 .0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm; and wherein if theurine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, clinical information comprising gender of the renal transplant recipient, age of the renal transplant recipient at time of transplant of the kidney graft, plasma level of BK polyomavirus (BKV) in the renal transplant recipient, level of serum creatinine in the renal transplant recipient, and estimated glomerular filtration rate (eGFR) of the in the renal transplant recipient in combination with the intensities of one or more differential NMR metabolite features of the one or more metabolites in the urine sample are used to classify the kidney graft as stable or unstable at risk of injury using a second machine learning model, wherein the one or more differential NMR metabolite features are selected from an NMR metabolite feature having a chemical shift at 4.261H ppm and a chemical shift at 68.9913C ppm, an NMR metabolite feature having a chemical shift at 3.871H ppm and a chemical shift at 58.813C ppm, an NMR metabolite feature having a chemical shift at 7.141H ppm and a chemical shift at 119.9313C ppm, an NMR metabolite feature having a chemical shift at 3.22 ’H ppm and a chemical shift at 30.1213C ppm, an NMR metabolite feature having a chemical shift at 8.181H ppm and a chemical shift at 137.9213C ppm, and one or more NMR metabolite features for one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine; and (c) displaying information regarding whether the kidney graft is stable or unstable at risk of injury.

[0280] In certain embodiments, wherein the kidney graft is classified as unstable at risk of injury, the method further comprises determining whether the renal transplant recipient is underimmunosuppressed or over-immunosuppressed, wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, clinical information comprising the age of the renal transplant recipient at the time of transplant of the kidney graft and the plasma level of BKV in the renal transplant recipient in combination with the intensities of differential NMR metabolite features comprising X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm and X2473 having a chemical shift at 3.811H ppm and a chemical shift at 63.9313C ppm are used to classify the renal transplant recipient as under-immunosuppressed or over-immunosuppressed using a third machine learning model, and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, clinical information comprising the gender of the renal transplant recipient, the age of the renal transplant recipient at time of transplant of the kidney graft, the plasma level of BKV in the renal transplant recipient, the level of serum creatinine in the renaltransplant recipient, and the eGFR of the in the renal transplant recipient in combination with the intensities of the one or more differential NMR metabolite features selected from the NMR metabolite feature having the chemical shift at 4.261H ppm and the chemical shift at 68.9913C ppm, the NMR metabolite feature having the chemical shift at 3.871H ppm and the chemical shift at 58.813C ppm, the NMR metabolite feature having the chemical shift at 7.141H ppm and the chemical shift at 119.93 13C ppm, the NMR metabolite feature having the chemical shift at 3.221H ppm and the chemical shift at 30.1213C ppm, the NMR metabolite feature having the chemical shift at 8.181H ppm and the chemical shift at 137.9213C ppm, and the one or more NMR metabolite features for the one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine are used to classify the renal transplant recipient as under-immunosuppressed or overimmunosuppressed using a fourth machine learning model; and displaying information regarding whether the renal transplant recipient is under-immunosuppressed or over-immunosuppressed.

[0281] The methods can be implemented in digital electronic circuitry, or in computer software, firmware, or hardware. The disclosed and other embodiments can be implemented as one or more computer program products, i.e., one or more modules of computer program instructions encoded on a computer readable medium for execution by, or to control the operation of, a data processing apparatus. The computer readable medium can be a machine-readable storage device, a machine- readable storage substrate, a memory device, a composition of matter effecting a machine-readable propagated signal, or any combination thereof.

[0282] A computer program (also known as a program, software, software application, script, or code) can be written in any form of programming language, including compiled or interpreted languages, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. A computer program does not necessarily correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub programs, or portions of code). A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network.

[0283] In a further aspect, the system for performing the computer implemented method, as described, may include a processor, a storage component (i.e., memory), a display component, andother components typically present in general purpose computers. In some embodiments, the processor is provided by a computer or handheld device (e.g., a cell phone or tablet). The storage component stores information accessible by the processor, including instructions that may be executed by the processor and data that may be retrieved, manipulated or stored by the processor.

[0284] The storage component includes instructions. For example, the storage component includes instructions for evaluating whether a renal transplant recipient is over-immunosuppressed or underimmunosuppressed according to the methods described herein. The computer processor is coupled to the storage component and configured to execute the instructions stored in the storage component in order to receive NMR and / or LC-MS spectra of one or more metabolites from a urine sample obtained from the renal transplant recipient and analyze the spectra according to one or more algorithms, as described herein (see, e.g., Examples).

[0285] The processor and / or memory may be operably connected to a display device, for example, via a wired, such as a Universal Serial Bus (USB) connection, or wireless connection, such as a Bluetooth connection. Any convenient display device, such as a liquid crystal display (LCD), lightemitting diode (LED) display, plasma (PDP) display, quantum dot (QLED) display or cathode ray tube display device may be used. The display component displays information regarding whether the renal transplant recipient is over-immunosuppressed or under-immunosuppressed.

[0286] The storage component may be of any type capable of storing information accessible by the processor, such as a hard-drive, memory card, ROM, RAM, DVD, CD-ROM, USB Flash drive, write- capable, and read-only memories. The processor may be a general purpose processor, a graphics processor unit, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor can be a microprocessor, but in the alternative, the processor can be a controller, microcontroller, or state machine, combinations of the same, or the like. A processor can also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. Although described herein primarily with respect to digital technology, a processor can also include primarily analog components. A computing environment can include any type of computer system, including, but not limited to, a computer system based on a microprocessor, a graphics processor unit, a mainframe computer, a digital signal processor, a portable computing device, a personal organizer, a device controller, and a computational engine within an appliance, to name a few.

[0287] The steps of a method, process, or algorithm described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module, engine, and associated databases can reside in memory resources such as in RAM memory, FRAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of non-transitory computer-readable storage medium, media, or physical computer storage known in the art. An exemplary storage medium can be coupled to the processor such that the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium can be integral to the processor. The processor and the storage medium can reside in an ASIC. The ASIC can reside in a user terminal. In the alternative, the processor and the storage medium can reside as discrete components in a user terminal.

[0288] The instructions may be any set of instructions to be executed directly (such as machine code) or indirectly (such as scripts) by the processor. In that regard, the terms "instructions," "steps" and "programs" may be used interchangeably herein. The instructions may be stored in object code form for direct processing by the processor, or in any other computer language including scripts or collections of independent source code modules that are interpreted on demand or compiled in advance.

[0289] Data may be retrieved, stored or modified by the processor in accordance with the instructions. For instance, although the system is not limited by any particular data structure, the data may be stored in computer registers, in a relational database as a table having a plurality of different fields and records, XML documents, or flat files. The data may also be formatted in any computer-readable format such as, but not limited to, binary values, ASCII or Unicode. Moreover, the data may comprise any information sufficient to identify the relevant information, such as numbers, descriptive text, proprietary codes, pointers, references to data stored in other memories (including other network locations) or information which is used by a function to calculate the relevant data.

[0290] In certain embodiments, the processor and storage component may comprise multiple processors and storage components that may or may not be stored within the same physical housing. For example, some of the instructions and data may be stored on removable CD-ROM and others within a read-only computer chip. Some or all of the instructions and data may be stored in a location physically remote from, yet still accessible by, the processor. Similarly, the processor may comprise a collection of processors which may or may not operate in parallel.

[0291] In some embodiments, the method can be performed using a cloud computing system. In these embodiments, the NMR and / or LC-MS spectra of one or more metabolites from a urine sampleobtained from a renal transplant recipient and the programming can be exported to a cloud computer, which runs the program, and returns an output to the user.Kits

[0292] Also provided are kits for use in performing the methods described herein for detecting overimmunosuppression or under-immunosuppression in a renal transplant recipient undergoing immunosuppressive therapy. The kit may comprise a container for holding a urine sample collected from a renal transplant recipient. The kit may further comprise one or more control reference samples or standards for performing NMR or LC-MS. The kit may also comprise one or more containers such as bottles, vials, syringes, and test tubes. Containers can be formed from a variety of materials, including glass or plastic.

[0293] In certain embodiments, the kit comprises software for carrying out the computer implemented methods, described herein, for evaluating whether a renal transplant recipient is stable or unstable at risk of injury, over-immunosuppressed, or under-immunosuppressed. In some embodiments, the kit comprises a non-transitory computer-readable medium comprising program instructions that, when executed by a processor in a computer, causes the processor to perform a computer implemented method described herein. In some embodiments, the kit comprises a system comprising a processor programmed to determine whether a renal transplant recipient is overimmunosuppressed or under-immunosuppressed according to a computer implemented method described herein; and a display component for displaying information regarding whether the renal transplant recipient is over-immunosuppressed or under-immunosuppressed.

[0294] In addition to the above components, the subject kits may further include (in certain embodiments) instructions for practicing the subject methods. For example, the kit may include instructions for classifying a renal transplant recipient as over-immunosuppressed, underimmunosuppressed, or stable using the methods described herein. These instructions may be present in the subject kits in a variety of forms, one or more of which may be present in the kit. One form in which these instructions may be present is as printed information on a suitable medium or substrate, e.g., a piece or pieces of paper on which the information is printed, in the packaging of the kit, in a package insert, and the like. Yet another form of these instructions is a computer readable medium, e.g., diskette, compact disk (CD), DVD, Blu-ray, flash drive, and the like, on which the information has been recorded. Yet another form of these instructions that may be present is a website address which may be used via the internet to access the information at a removed site. The kit may further include instructions to halt, alter, or monitor a patient's treatment based on an analysis of metabolite levels in a urine sample from the patient according to the methods described herein.Examples of Non-Limiting Aspects of the Disclosure

[0295] Aspects, including embodiments, of the present subject matter described above may be beneficial alone or in combination, with one or more other aspects or embodiments. Without limiting the foregoing description, certain non-limiting aspects of the disclosure numbered 1 -141 are provided below. As will be apparent to those of skill in the art upon reading this disclosure, each of the individually numbered aspects may be used or combined with any of the preceding or following individually numbered aspects. This is intended to provide support for all such combinations of aspects and is not limited to combinations of aspects explicitly provided below:1. A method of monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy, the method comprising:(a) obtaining a urine sample from the renal transplant recipient; and(b) measuring levels of one or more metabolites in the urine sample using nuclear magnetic resonance (NMR), wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, intensities of one or more differential NMR metabolite features of one or more metabolites in the urine sample are used to classify the kidney graft as stable or unstable at risk of injury using a first machine learning model, wherein the one or more differential NMR metabolite features are selected from X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.97 1H ppm and a chemical shift at 46.5513C ppm; and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, clinical information comprising gender of the renal transplant recipient, age of the renal transplant recipient at time of transplant of the kidney graft, plasma level of BK polyomavirus (BKV) in the renal transplant recipient, level of serum creatinine in the renal transplant recipient, and estimated glomerular filtration rate (eGFR) of the renal transplant recipient in combination with intensities of one or more differential NMR metabolite features of one or more metabolites in the urine sample are usedto classify the kidney graft as stable or unstable at risk of injury using a second machine learning model, wherein the one or more differential NMR metabolite features are selected from an NMR metabolite feature having a chemical shift at 4.261H ppm and a chemical shift at 68.9913C ppm, an NMR metabolite feature having a chemical shift at 3.871H ppm and a chemical shift at 58.813C ppm, an NMR metabolite feature having a chemical shift at 7.141H ppm and a chemical shift at 119.9313C ppm, an NMR metabolite feature having a chemical shift at 3.221H ppm and a chemical shift at 30.1213C ppm, an NMR metabolite feature having a chemical shift at 8.181H ppm and a chemical shift at 137.9213C ppm, and one or more NMR metabolite features for one or more metabolites selected from 4-hydroxy-3- methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine.2. The method of aspect 1 , wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, the one or more differential NMR metabolite features comprise the X65 having the chemical shift at 4.061H ppm and the chemical shift at 85.9413C ppm, the X56 having the chemical shift at 4.671H ppm and the chemical shift at 98.5713C ppm, the X360 having the chemical shift at 2.171H ppm and the chemical shift at 24.5113C ppm, the X333 having the chemical shift at 4.441H ppm and the chemical shift at 51.0413C ppm, the X93 having the chemical shift at 4.081H ppm and the chemical shift at 74.8313C ppm, the X129 having the chemical shift at 3.281H ppm and the chemical shift at 62.2313C ppm, and the X161 having the chemical shift at 3.971H ppm and the chemical shift at 46.5513C ppm; and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, the one or more differential NMR metabolite features comprise the NMR metabolite feature having the chemical shift at 4.261H ppm and the chemical shift at 68.9913C ppm, the NMR metabolite feature having the chemical shift at 3.871H ppm and the chemical shift at 58.813C ppm, the NMR metabolite feature having the chemical shift at 7.141H ppm and a chemical shift at 119.9313C ppm, the NMR metabolite feature havingthe chemical shift at 3.22 ’H ppm and the chemical shift at 30.1213C ppm, the NMR metabolite feature having the chemical shift at 8.181H ppm and the chemical shift at 137.9213C ppm, and the one or more NMR metabolite features for the one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine.3. The method of aspect 1 or 2, wherein if the kidney graft is classified as unstable at risk of injury, the method further comprises determining whether the renal transplant recipient is under-immunosuppressed or over-immunosuppressed, wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, clinical information comprising the age of the renal transplant recipient at the time of transplant of the kidney graft and the plasma level of BKV in the renal transplant recipient in combination with intensities of differential NMR metabolite features comprising X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm and X2473 having a chemical shift at 3.811H ppm and a chemical shift at 63.9313C ppm are used to classify the renal transplant recipient as under-immunosuppressed or overimmunosuppressed using a third machine learning model, and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, clinical information comprising the gender of the renal transplant recipient, the age of the renal transplant recipient at time of transplant of the kidney graft, the plasma level of BKV in the renal transplant recipient, the level of serum creatinine in the renal transplant recipient, and the eGFR of the renal transplant recipient in combination with intensities of the one or more differential NMR metabolite features selected from the NMR metabolite feature having the chemical shift at 4.261H ppm and the chemical shift at 68.9913C ppm, the NMR metabolite feature having the chemical shift at 3.871H ppm and the chemical shift at 58.813C ppm, the NMR metabolite feature having the chemical shift at 7.141H ppm and the chemical shift at 1 19.9313C ppm, the NMR metabolite feature having the chemical shift at 3.221H ppm and the chemical shift at 30.1213C ppm, the NMR metabolite feature having the chemical shift at 8.181H ppm and the chemical shift at 137.9213C ppm, and the one or more NMR metabolite features for the one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine are used to classify the renal transplant recipient as under-immunosuppressed or overimmunosuppressed using a fourth machine learning model.4. The method of aspect 3, further comprising altering the immunosuppressive therapy administered to the renal transplant recipient if the renal transplant recipient is determined to be under-immunosuppressed or over-immunosuppressed, wherein immunosuppression is increased if the renal transplant recipient is determined to be under-immunosuppressed and decreased if the renal transplant recipient is determined to be over-immunosuppressed.5. The method of aspect 3 or 4, further comprising performing diagnostic screening to determine whether the renal transplant recipient has an infection, a malignancy, or polyomavirus- associated nephropathy if the renal transplant recipient is determined to be overimmunosuppressed; and treating the renal transplant recipient for the infection, the malignancy, or the polyomavirus-associated nephropathy if the renal transplant recipient is diagnosed as having the infection, the malignancy, or the polyomavirus-associated nephropathy.6. The method of aspect 3 or 4, further comprising performing diagnostic screening to determine whether the renal transplant recipient has graft dysfunction, subclinical graft rejection, or graft rejection if the renal transplant recipient is determined to be under-immunosuppressed; and treating the renal transplant recipient for the graft dysfunction, subclinical graft rejection, or graft rejection if the renal transplant recipient is diagnosed as having the graft dysfunction, subclinical graft rejection, or graft rejection.7. The method of any one of aspects 1 -6, wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, the one or metabolites comprise glucuronic acid, trigonelline, trimethylamine N- oxide (TMAO), and hippuric acid.8. The method of any one of aspects 1 -7, wherein the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA) or a random forest (RF) machine learning algorithm.9. The method of any one of aspects 1 -8, wherein the NMR spectroscopy comprises one-dimensional (1 D)1H NMR spectroscopy, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectroscopy, or a combination thereof.10. The method of aspect 9, wherein the NMR is performed with non-uniformed sampling(NUS).11. The method of any one of aspects 1 -10, further comprising using one or more normalization NMR metabolite features.12. A computer implemented method for monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy, the computer performing steps comprising:(a) receiving nuclear magnetic resonance (NMR) spectra of one or more metabolites from a urine sample obtained from the renal transplant recipient;(b) measuring intensities of one or more differential NMR metabolite features in the NMR spectra, wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, the intensities of the one or more differential NMR metabolite features of the one or more metabolites in the urine sample are used to classify the kidney graft as stable or unstable at risk of injury using a first machine learning model, wherein the one or more differential NMR metabolite features are selected from X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.94 13C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm; and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, clinical information comprising gender of the renal transplant recipient, age of the renal transplant recipient at time of transplant of the kidney graft, plasma level of BK polyomavirus (BKV) in the renal transplant recipient, level of serum creatinine in the renal transplant recipient, and estimated glomerular filtration rate (eGFR) of the renal transplant recipient in combination with the intensities of one or more differential NMR metabolite features of the one or more metabolites in the urine sample are used to classify the kidney graft as stable or unstable at risk of injury using a second machine learning model, wherein the one or more differential NMR metabolite features are selectedfrom an NMR metabolite feature having a chemical shift at 4.261H ppm and a chemical shift at 68.9913C ppm, an NMR metabolite feature having a chemical shift at 3.871H ppm and a chemical shift at 58.813C ppm, an NMR metabolite feature having a chemical shift at 7.141H ppm and a chemical shift at 119.9313C ppm, an NMR metabolite feature having a chemical shift at 3.221H ppm and a chemical shift at 30.1213C ppm, an NMR metabolite feature having a chemical shift at 8.181H ppm and a chemical shift at 137.9213C ppm, and one or more NMR metabolite features for one or more metabolites selected from 4-hydroxy-3- methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine; and (c) displaying information regarding whether the kidney graft is stable or unstable at risk of injury.13. The computer implemented method of aspect 12, wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, the one or more differential NMR metabolite features comprise the X65 having the chemical shift at 4.061H ppm and the chemical shift at 85.9413C ppm, the X56 having the chemical shift at 4.67 ’H ppm and the chemical shift at 98.5713C ppm, the X360 having the chemical shift at 2.171H ppm and the chemical shift at 24.5113C ppm, the X333 having the chemical shift at 4.441H ppm and the chemical shift at 51.0413C ppm, the X93 having the chemical shift at 4.081H ppm and the chemical shift at 74.8313C ppm, the X129 having the chemical shift at 3.281H ppm and the chemical shift at 62.2313C ppm, and the X161 having the chemical shift at 3.971H ppm and the chemical shift at 46.5513C ppm; and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, the one or more differential NMR metabolite features comprise the NMR metabolite feature having the chemical shift at 4.261H ppm and the chemical shift at 68.9913C ppm, the NMR metabolite feature having the chemical shift at 3.871H ppm and the chemical shift at 58.813C ppm, the NMR metabolite feature having the chemical shift at 7.141H ppm and a chemical shift at 119.9313C ppm, the NMR metabolite feature havingthe chemical shift at 3.22 ’H ppm and the chemical shift at 30.1213C ppm, the NMR metabolite feature having the chemical shift at 8.181H ppm and the chemical shift at 137.9213C ppm, and the one or more NMR metabolite features for the one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine.14. The computer implemented method of aspect 12 or 13, wherein if the kidney graft is classified as unstable at risk of injury, the method further comprises determining whether the renal transplant recipient is under-immunosuppressed or over-immunosuppressed, wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, clinical information comprising the age of the renal transplant recipient at the time of transplant of the kidney graft and the plasma level of BKV in the renal transplant recipient in combination with the intensities of differential NMR metabolite features comprising X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm and X2473 having a chemical shift at 3.811H ppm and a chemical shift at 63.9313C ppm are used to classify the renal transplant recipient as under-immunosuppressed or overimmunosuppressed using a third machine learning model, and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, clinical information comprising the gender of the renal transplant recipient, the age of the renal transplant recipient at time of transplant of the kidney graft, the plasma level of BKV in the renal transplant recipient, the level of serum creatinine in the renal transplant recipient, and the eGFR of the renal transplant recipient in combination with the intensities of the one or more differential NMR metabolite features selected from the NMR metabolite feature having the chemical shift at 4.261H ppm and the chemical shift at 68.9913C ppm, the NMR metabolite feature having the chemical shift at 3.871H ppm and the chemical shift at 58.813C ppm, the NMR metabolite feature having the chemical shift at 7.141H ppm and the chemical shift at 1 19.9313C ppm, the NMR metabolite feature having the chemical shift at 3.221H ppm and the chemical shift at 30.1213C ppm, the NMR metabolite feature having the chemical shift at 8.181H ppm and the chemical shift at 137.9213C ppm, and the one or more NMR metabolite features for the one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine are used to classify the renal transplant recipient as under-immunosuppressed or overimmunosuppressed using a fourth machine learning model; anddisplaying information regarding whether the renal transplant recipient is underimmunosuppressed or over-immunosuppressed.15. The computer implemented method of any one of aspects 12-14, further comprising storing the information regarding whether the renal transplant recipient is under-immunosuppressed or over-immunosuppressed in a database.16. The computer implemented method of any one of aspects 12-15, wherein the NMR spectra comprise one-dimensional (1 D)1H NMR spectra, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectra, or a combination thereof.17. The computer implemented method of aspect 16, wherein the NMR spectra are obtained using non-uniformed sampling (NUS).18. A system for monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy using the computer implemented method of any one of aspects 12-17, the system comprising:(a) a storage component for storing data, wherein the storage component has instructions for monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy based on analysis of the NMR spectra of the one or more metabolites stored therein;(b) a computer processor programmed to analyze the NMR spectra using one or more algorithms, wherein the computer processor is coupled to the storage component and configured to execute the instructions stored in the storage component in order to receive the inputted NMR spectra, and analyze the NMR spectra according to the computer implemented method of any one of aspects 12-17; and(c) a display component for displaying information regarding whether the renal transplant recipient is stable, unstable at risk of injury, under-immunosuppressed, or over-immunosuppressed.19. A non-transitory computer-readable medium comprising program instructions that, when executed by a processor in a computer, causes the processor to perform the computer implemented method of any one of aspects 12-17.20. A kit comprising the non-transitory computer-readable medium of aspect 19 and instructions for monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy.21. A method of detecting over-immunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy, the method comprising:(a) obtaining a urine sample from the renal transplant recipient;(b) measuring levels of one or more metabolites in the urine; and(c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed.22. The method of aspect 21 , further comprising altering the immunosuppressive therapy administered to the renal transplant recipient if the renal transplant recipient is determined to be over-immunosuppressed.23. The method of aspect 21 or 22, further comprising performing diagnostic screening to determine whether the renal transplant recipient has an infection, a malignancy, or polyomavirus- associated nephropathy if the renal transplant recipient is determined to be overimmunosuppressed; and treating the renal transplant recipient for the infection, the malignancy, or the polyomavirus-associated nephropathy if the renal transplant recipient is diagnosed as having the infection, the malignancy, or the polyomavirus-associated nephropathy.24. The method of aspect 23, wherein the infection is a BK virus infection.25. The method of any one of aspects 21 -24, wherein said measuring comprises performing nuclear magnetic resonance (NMR) spectroscopy or mass spectrometry.26. The method of aspect 25, wherein the NMR spectroscopy is multi-dimensional NMR spectroscopy.27. The method of aspect 25, wherein the NMR spectroscopy comprises one-dimensional (1 D)1H NMR spectroscopy, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectroscopy, or a combination thereof.28. The method of aspect 26 or 27, wherein the NMR is performed with non-uniformed sampling (NUS).29. The method of any one of aspects 25-28, wherein one or more differential NMR metabolite features are used to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model, wherein the one or more differential NMR metabolite features are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.30. The method of aspect 29, wherein the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.28 ’H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.06 ’H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.31 . The method of aspect 29, wherein the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.32. The method of any one of aspects 25-31 , further comprising using one or more normalization NMR metabolite features in combination with the differential NMR metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.33. The method of aspect 32, wherein the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.56 ’H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.34. The method of aspect 33, wherein the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1 .561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.35. The method of aspect 32, wherein the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N- oxide, and glutamine NMR features.36. The method of any one of aspects 25-35, wherein chemical shifts are calibrated against an internal deuterated sodium 2,2-dimethyl-2-silapentane-5-sulfonate standard.37. The method of any one of aspects 21 -36, wherein said measuring comprises performing liquid chromatography-mass spectrometry (LC-MS).38. The method of aspect 37, wherein ionization is performed using heated electrospray ionization (HESI).39. The method of aspect 37 or 38, wherein the liquid chromatography is performed using a linear gradient of 5% to 95% acetonitrile.40. The method of any one of aspects 37-39, wherein one or more differential LC-MS metabolite features of caffeine, benzoyl formic acid, and cytidine are used to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.41 . The method of aspect 40, wherein the differential LC-MS metabolite features for caffeine comprise a feature with a mass-to-charge ratio (m / z) of 195.00 for a caffeine ([M+H]+ ion and a feature with a m / z of 137.90 and a feature with a m / z of 110.00 from selected reaction monitoring.42. The method of aspect 40 or 41 , wherein the differential LC-MS metabolite features for benzoyl formic acid comprise a feature with a m / z of 149.09 for a benzoyl formic acid [M-H]- ion and a feature with a m / z of 104.91 and a feature with a m / z of 77.07 from selected reaction monitoring.43. The method of any one of aspects 40-42, wherein the differential LC-MS metabolite features for cytidine comprise a feature with a m / z of 244.09 for a cytidine [M+H]+ ion and a feature with a m / z of 111 .90 from selected reaction monitoring.44. The method of any one of aspects 40-43, further comprising using one or more normalization LC-MS metabolite features in combination with the differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.45. The method of aspect 44, wherein the normalization LC-MS metabolite features comprise lysine, hippuric acid, mannitol, trimethylamine N-oxide, and glutamine LC-MS features.46. The method of any one of aspects 40-45, wherein one or more differential NMR metabolite features are used in combination with one or more differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model,wherein the one or more differential NMR metabolite features are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm; and wherein the one or more differential LC-MS metabolite features are selected from a feature with a mass-to-charge ratio (m / z) of 195.00 for a caffeine ([M+H]+ ion and a feature with a m / z of 137.90 and a feature with a m / z of 110.00 from selected reaction monitoring, a feature with a m / z of 149.09 for a benzoyl formic acid [M-H]- ion and a feature with a m / z of 104.91 and a feature with a m / z of 77.07 from selected reaction monitoring, a feature with a m / z of 244.09 for a cytidine [M+H]+ ion and a feature with a m / z of 111 .90 from selected reaction monitoring.47. The method of aspect 46, wherein the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.48. The method of aspect 46, wherein the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.49. The method of any one of aspects 46-48, further comprising using one or more normalization NMR metabolite features and normalization LC-MS metabolite features in combination with the differential NMR metabolite features and differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.50. The method of aspect 49, wherein the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.82 ’H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.56 ’H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.51. The method of aspect 50, wherein the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1 .561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.52. The method of aspect 49, wherein the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N- oxide, and glutamine NMR features.53. The method of any one of aspects 49-52, wherein the normalization LC-MS metabolite features comprise lysine, hippuric acid, mannitol, trimethylamine N-oxide, and glutamine LC-MS features.54. The method of any one of aspects 21-53, further comprising isolating metabolites from the urine sample prior to said measuring.55. The method of any one of aspects 21-54, wherein the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA) or a random forest (RF) machine learning algorithm.56. The method of any one of aspects 21-55, further comprising repeating steps (a) - (c) periodically to determine whether the renal transplant recipient has become overimmunosuppressed.57. The method of aspect 56, wherein steps (a) - (c) are repeated at least twice a month, at least once a month, at least every 2 months, at least every 3 months, at least every 4 months, at least every 5 months, at least every 6 months, or at least once a year.58. The method of any one of aspects 21 -57, wherein the one or more metabolites are selected from lysine, caffeine, benzoyl formic acid, and cytidine.59. A biomarker selected from lysine, caffeine, benzoyl formic acid, and cytidine for use in a method of diagnosing over-immunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy.60. A kit comprising: a container for collecting a urine sample from a renal transplant recipient, and instructions to halt, alter, or monitor treatment of the renal transplant recipient based on analyzing the levels of one or more metabolites in the urine sample using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed according to the method of any one of aspects 21-58.61 . A computer implemented method for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed, the computer performing steps comprising:(a) receiving nuclear magnetic resonance (NMR) and / or liquid chromatography-mass spectrometry (LC-MS) spectra of one or more metabolites from a urine sample obtained from the renal transplant recipient;(b) measuring levels of the one or more metabolites using the NMR and / or LC-MS spectra;(c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed; and(d) displaying information regarding whether the renal transplant recipient is overimmunosuppressed.62. The computer implemented method of aspect 61 , further comprising storing the information regarding whether the renal transplant recipient is over-immunosuppressed in a database.63. The computer implemented method of aspect 61 or 62, wherein the NMR spectra are multi-dimensional NMR spectra.64. The computer implemented method of aspect 63, wherein the NMR spectra comprise one-dimensional (1 D)1H NMR spectra, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectra, or a combination thereof.65. The computer implemented method of aspect 63 or 64, wherein the NMR spectra are obtained using non-uniformed sampling (NUS).66. The computer implemented method of any one of aspects 61 -65, wherein one or more differential NMR metabolite features in the NMR spectra are used to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model, wherein the one or more differential NMR metabolite features are selected from X109 having a chemical shift at 3.01 ’H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.67. The computer implemented method of aspect 66, wherein the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having achemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.68. The computer implemented method of aspect 66, wherein the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.69. The computer implemented method of any one of aspects 61 -68, further comprising using one or more normalization NMR metabolite features in combination with the differential NMR metabolite features in the NMR spectra to classify the renal transplant recipient as overimmunosuppressed or stable using the machine learning model.70. The computer implemented method of aspect 69, wherein the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1 .561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.71. The computer implemented method of aspect 70, wherein the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppmand a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.72. The computer implemented method of aspect 69, wherein the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.73. The computer implemented method of any one of aspects 61 -72, wherein chemical shifts are calibrated against an internal deuterated sodium 2,2-dimethyl-2-silapentane-5-sulfonate standard.74. The computer implemented method of any one of aspects 61 -73, wherein one or more differential LC-MS metabolite features of caffeine, benzoyl formic acid, and cytidine in the LC-MS spectra are used to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.75. The computer implemented method of aspect 74, wherein the differential LC-MS metabolite features for caffeine comprise a feature with a mass-to-charge ratio (m / z) of 195.00 for a caffeine ([M+H]+ ion and a feature with a m / z of 137.90 and a feature with a m / z of 110.00 from selected reaction monitoring.76. The computer implemented method of aspect 74 or 75, wherein the differential LC- MS metabolite features for benzoyl formic acid comprise a feature with a m / z of 149.09 for a benzoyl formic acid [M-H]- ion and a feature with a m / z of 104.91 and a feature with a m / z of 77.07 from selected reaction monitoring.77. The computer implemented method of any one of aspects 74-76, wherein the differential LC-MS metabolite features for cytidine comprise a feature with a m / z of 244.09 for a cytidine [M+H]+ ion and a feature with a m / z of 111 .90 from selected reaction monitoring.78. The computer implemented method of any one of aspects 74-77, further comprising using one or more normalization LC-MS metabolite features in combination with the differential LC-MS metabolite features in the LC-MS spectra to classify the renal transplant recipient as overimmunosuppressed or stable using the machine learning model.79. The computer implemented method of aspect 78, wherein the normalization LC-MS metabolite features comprise lysine, hippuric acid, mannitol, trimethylamine N-oxide, and glutamine LC-MS features.80. The computer implemented method of any one of aspects 61 -79, wherein one or more differential NMR metabolite features are used in combination with one or more differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model, wherein the one or more differential NMR metabolite features are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm; and wherein the one or more differential LC-MS metabolite features are selected from a feature with a mass-to-charge ratio (m / z) of 195.00 for a caffeine ([M+H]+ ion and a feature with a m / z of 137.90 and a feature with a m / z of 110.00 from selected reaction monitoring, a feature with a m / z of 149.09 for a benzoyl formic acid [M-H]- ion and a feature with a m / z of 104.91 and a feature with a m / z of 77.07 from selected reaction monitoring, a feature with a m / z of 244.09 for a cytidine [M+H]+ ion and a feature with a m / z of 111 .90 from selected reaction monitoring.81 . The computer implemented method of aspect 80, wherein the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.82. The computer implemented method of aspect 80, wherein the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.83. The computer implemented method of any one of aspects 80-82, further comprising using one or more normalization NMR metabolite features and normalization LC-MS metabolite features in combination with the differential NMR metabolite features and differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.84. The computer implemented method of aspect 83, wherein the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1 .561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.85. The computer implemented method of aspect 84, wherein the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemicalshift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.86. The computer implemented method of aspect 83, wherein the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.87. The computer implemented method of any one of aspects 83-86, wherein the normalization LC-MS metabolite features comprise lysine, hippuric acid, mannitol, trimethylamine N- oxide, and glutamine LC-MS features.88. The computer implemented method of any one of aspects 61 -87, wherein the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA) or a random forest (RF) machine learning algorithm.89. A system for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed using the computer implemented method of any one of aspects 61-88, the system comprising:(a) a storage component for storing data, wherein the storage component has instructions for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed based on analysis of the NMR and / or LC-MS spectra of the one or more metabolites stored therein;(b) a computer processor programmed to analyze the NMR and / or LC-MS spectra using one or more algorithms, wherein the computer processor is coupled to the storage component and configured to execute the instructions stored in the storage component in order to receive the inputted NMR and / or LC-MS spectra, and analyze the NMR and / or LC-MS spectra according to the computer implemented method of any one of aspects 61-88; and(c) a display component for displaying information regarding whether the renal transplant recipient is over-immunosuppressed.90. A non-transitory computer-readable medium comprising program instructions that, when executed by a processor in a computer, causes the processor to perform the computer implemented method of any one of aspects 61 -88.91. A kit comprising the non-transitory computer-readable medium of aspect 90 and instructions for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed.92. A method of detecting over-immunosuppression or under-immunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy, the method comprising:(a) obtaining a urine sample from the renal transplant recipient;(b) measuring levels of one or more metabolites in the urine; and(c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed or underimmunosuppressed.93. The method of aspect 92, wherein the one or more metabolites are selected from lysine, caffeine, benzoyl formic acid, cytidine, glucuronic acid, glucaric acid, lactose, glucose-6- phosphate, 1 -methylhistamine, histidine, trigonelline, trimethylamine N-oxide (TMAO), hippuric acid, choline, phosphoethanolamine, phosphocholine, acetylglycine, phenyllactic acid, mycophenolic acid glucuronide (MPAG), creatine, and hydroxyphenylacetic acid.94. The method of aspect 92 or 93, further comprising altering the immunosuppressive therapy administered to the renal transplant recipient if the renal transplant recipient is determined to be over-immunosuppressed or under-immunosuppressed, wherein immunosuppression is decreased if the renal transplant recipient is determined to be over-immunosuppressed and increased if the renal transplant recipient is determined to be under-immunosuppressed.95. The method of any one of aspects 92-94, further comprising performing diagnostic screening to determine whether the renal transplant recipient has an infection, a malignancy, or polyomavirus-associated nephropathy if the renal transplant recipient is determined to be overimmunosuppressed; and treating the renal transplant recipient for the infection, the malignancy, or the polyomavirus-associated nephropathy if the renal transplant recipient is diagnosed as having the infection, the malignancy, or the polyomavirus-associated nephropathy.96. The method of any one of aspects 92-94, further comprising performing diagnostic screening to determine whether the renal transplant recipient has graft dysfunction, subclinical graftrejection, or graft rejection if the renal transplant recipient is determined to be underimmunosuppressed; and treating the renal transplant recipient for the graft dysfunction, subclinical graft rejection, or graft rejection.97. The method of any one of aspects 92-96, wherein the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA) or a random forest (RF) machine learning algorithm.98. The method of any one of aspects 92-97, wherein said measuring comprises performing nuclear magnetic resonance (NMR) spectroscopy or mass spectrometry.99. The method of aspect 98, wherein the NMR spectroscopy is multi-dimensional NMR spectroscopy.100. The method of aspect 99, wherein the NMR spectroscopy comprises one-dimensional (1 D)1H NMR spectroscopy, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectroscopy, or a combination thereof.101. The method of aspect 99 or 100, wherein the NMR is performed with non-uniformed sampling (NUS).102. The method of any one of aspects 98-101 , wherein one or more differential NMR metabolite features are used to classify the renal transplant recipient as over-immunosuppressed, under-immunosuppressed, or stable using the machine learning model, wherein the one or more differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm; and wherein the one or more differential NMR metabolite features used to classify the renal transplant recipient as under-immunosuppressed or stable are selected from X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671Hppm and a chemical shift at 98.5713C ppm, X1223 having a chemical shift at 7.941H ppm and a chemical shift at 140.3013C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.103. The method of aspect 102, wherein the differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.104. The method of aspect 102, wherein the differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable comprise or consist of X109 having a chemical shift at 3.01 ’H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.105. The method of any one of aspects 98-104, wherein the differential NMR metabolite features used to classify the renal transplant recipient as under-immunosuppressed or stable comprise or consist of X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.106. The method of any one of aspects 98-105, further comprising using one or more normalization NMR metabolite features in combination with the differential NMR metabolite featuresto classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.107. The method of aspect 106, wherein the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.108. The method of aspect 107, wherein the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1 .561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.109. The method of aspect 106, wherein the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N- oxide, and glutamine NMR features.110. The method of any one of aspects 98-109, wherein chemical shifts are calibrated against an internal deuterated sodium 2,2-dimethyl-2-silapentane-5-sulfonate standard.111. The method of any one of aspects 92-1 10, further comprising repeating steps (a) - (c) periodically to determine whether the renal transplant recipient has become overimmunosuppressed.112. The method of aspect 111 , wherein steps (a) - (c) are repeated at least twice a month, at least once a month, at least every 2 months, at least every 3 months, at least every 4 months, at least every 5 months, at least every 6 months, or at least once a year.113. A biomarker selected from lysine, caffeine, benzoyl formic acid, cytidine, glucuronic acid, glucaric acid, lactose, glucose-6-phosphate, 1 -methylhistamine, histidine, trigonelline, trimethylamine N-oxide (TMAO), hippuric acid, choline, phosphoethanolamine, phosphocholine, acetylglycine, phenyllactic acid, mycophenolic acid glucuronide (MPAG), creatine, and hydroxyphenylacetic acid for use in a method of diagnosing over-immunosuppression or underimmunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy.114. A kit comprising: a container for collecting a urine sample from a renal transplant recipient, and instructions to halt, alter, or monitor treatment of the renal transplant recipient based on analyzing the levels of one or more metabolites in the urine sample using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed or underimmunosuppressed according to the method of any one of aspects 92-112.115. A computer implemented method for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed, underimmunosuppressed, or stable, the computer performing steps comprising:(a) receiving nuclear magnetic resonance (NMR) and / or mass spectrometry spectra of one or more metabolites from a urine sample obtained from the renal transplant recipient;(b) measuring levels of the one or more metabolites using the NMR spectra and / or mass spectrometry spectra;(c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed, underimmunosuppressed, or stable; and(d) displaying information regarding whether the renal transplant recipient is overimmunosuppressed, under-immunosuppressed, or stable.116. The computer implemented method of aspect 115, further comprising storing the information regarding whether the renal transplant recipient is over-immunosuppressed, underimmunosuppressed, or stable in a database.117. The computer implemented method of aspect 115 or 116, wherein the NMR spectra are multi-dimensional NMR spectra.118. The computer implemented method of aspect 117, wherein the NMR spectra comprise one-dimensional (1 D)1H NMR spectra, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectra, or a combination thereof.119. The computer implemented method of aspect 117 or 118, wherein the NMR spectra are obtained using non-uniformed sampling (NUS).120. The computer implemented method of any one of aspects 115-119, wherein one or more differential NMR metabolite features are used to classify the renal transplant recipient as overimmunosuppressed, under-immunosuppressed, or stable using the machine learning model, wherein the one or more differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm; and wherein the one or more differential NMR metabolite features used to classify the renal transplant recipient as under-immunosuppressed or stable are selected from X65 having a chemical shift at 4.06 ’H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X1223 having a chemical shift at 7.941H ppm and a chemical shift at 140.3013C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 havinga chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.121. The computer implemented method of aspect 120, wherein the differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.122. The computer implemented method of aspect 120, wherein the differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.123. The computer implemented method of any one of aspects 115-122, wherein the differential NMR metabolite features used to classify the renal transplant recipient as underimmunosuppressed or stable comprise or consist of X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51 .0413C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.124. The computer implemented method of any one of aspects 115-123, further comprising using one or more normalization NMR metabolite features in combination with the differential NMR metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.125. The computer implemented method of aspect 124, wherein the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1 .561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.126. The computer implemented method of aspect 125, wherein the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.24 ’H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.127 The computer implemented method of aspect 124, wherein the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.128. A system for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed or under-immunosuppressed using the computer implemented method of any one of aspects 115-127, the system comprising:(a) a storage component for storing data, wherein the storage component has instructions for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed or under-immunosuppressed based on analysis of the NMR spectra of the one or more metabolites stored therein;(b) a computer processor programmed to analyze the NMR spectra using one or more algorithms, wherein the computer processor is coupled to the storage component and configured to execute the instructions stored in the storage component in order to receive the inputted NMR spectra, and analyze the NMR spectra according to the computer implemented method of any one of aspects 95-107; and(c) a display component for displaying information regarding whether the renal transplant recipient is over-immunosuppressed or under-immunosuppressed.129. A non-transitory computer-readable medium comprising program instructions that, when executed by a processor in a computer, causes the processor to perform the computer implemented method of any one of aspects 115-127.130. A kit comprising the non-transitory computer-readable medium of aspect 129 and instructions for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed or under-immunosuppressed.131. A method of detecting under-immunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy, the method comprising:(a) obtaining a urine sample from the renal transplant recipient;(b) measuring levels of one or more metabolites in the urine; and(c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is under-immunosuppressed.132. The method of aspect 131 , wherein the one or more metabolites are selected from glucuronic acid, glucaric acid, lactose, glucose-6-phosphate, 1 -methylhistamine, histidine, trigonelline, trimethylamine N-oxide (TMAO), hippuric acid, choline, phosphoethanolamine, phosphocholine, acetylglycine, phenyllactic acid, mycophenolic acid glucuronide (MPAG), creatine, and hydroxyphenylacetic acid.133. The method of aspect 131 or 132, further comprising altering the immunosuppressive therapy administered to the renal transplant recipient if the renal transplant recipient is determined to be under-immunosuppressed, wherein immunosuppression is increased if the renal transplant recipient is determined to be under-immunosuppressed.134. The method of any one of aspects 131 -133, further comprising performing diagnostic screening to determine whether the renal transplant recipient has graft dysfunction, subclinical graft rejection, or graft rejection if the renal transplant recipient is determined to be underimmunosuppressed; and treating the renal transplant recipient for the graft dysfunction, subclinical graft rejection, or graft rejection.135. The method of any one of aspects 131-134, wherein the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA) or a random forest (RF) machine learning algorithm.136. The method of any one of aspects 131 -135, wherein said measuring comprises performing nuclear magnetic resonance (NMR) spectroscopy or mass spectrometry.137. The method of aspect 136, wherein the NMR spectroscopy is multi-dimensional NMR spectroscopy.138. The method of aspect 137, wherein the NMR spectroscopy comprises onedimensional (1 D)1H NMR spectroscopy, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectroscopy, or a combination thereof.139. The method of aspect 137 or 138, wherein the NMR is performed with non-uniformed sampling (NUS).140. The method of any one of aspects 136-139, wherein one or more differential NMR metabolite features are used to classify the renal transplant recipient as under-immunosuppressed or stable using the machine learning model, wherein the one or more differential NMR metabolite features are selected from X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X1223 having a chemical shift at 7.941H ppm and a chemical shift at 140.3013C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, X93 having a chemical shift at 4.08 ’H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.141. The method of aspect 140, wherein the differential NMR metabolite features used to classify the renal transplant recipient as under-immunosuppressed or stable comprise or consist of X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.

[0296] It will be apparent to one of ordinary skill in the art that various changes and modifications can be made without departing from the spirit or scope of the invention.EXPERIMENTAL

[0297] The following examples are put forth so as to provide those of ordinary skill in the art with a complete disclosure and description of how to make and use the present invention, and are not intended to limit the scope of what the inventors regard as their invention nor are they intended to represent that the experiments below are all or the only experiments performed. Efforts have been made to ensure accuracy with respect to numbers used (e.g. amounts, temperature, etc.) but some experimental errors and deviations should be accounted for. Unless indicated otherwise, parts are parts by weight, molecular weight is average molecular weight, temperature is in degrees Centigrade, and pressure is at or near atmospheric.

[0298] All publications and patent applications cited in this specification are herein incorporated by reference as if each individual publication or patent application were specifically and individually indicated to be incorporated by reference.

[0299] The present invention has been described in terms of particular embodiments found or proposed by the present inventor to comprise preferred modes for the practice of the invention. It will be appreciated by those of skill in the art that, in light of the present disclosure, numerous modifications and changes can be made in the particular embodiments exemplified without departing from the intended scope of the invention. All such modifications are intended to be included within the scope of the appended claims.Example 1Evaluation of Metabolite Signature to Mitigate Over-Immunosuppression in Stable Kidney T ransplant PatientsINTRODUCTION

[0300] Managing complications related to over-immunosuppression is a clinical challenge in posttransplant care, with infections accounting for the second leading cause of death with functioning graft (DWFG) in renal transplant recipients (RTRs) within the first year1. At present, there are no clinically validated biomarkers to detect over-immunosuppression2. Polyomavirus-associated nephropathy (PVAN) is the result of an opportunistic infection indicative of over-immunosuppression that occurs in 5-10% of RTRs, which can lead to graft dysfunction or loss3.

[0301] Metabolism is the set of biochemical processes that provides energy and biomass to sustain all cellular and biological processes. Metabolism and the immune system are intimately connected, wherein metabolite levels influence the immune response, and the immune system influences metabolite levels (Matarese, 2004). Previous reports demonstrated that immunosuppressant agents used in transplant elicit unique metabolic changes (Kim, 2010) and that distinct metabolic signatures are correlated with graft function (Dieme, 2014). These preliminary results suggest metabolites could represent a promising class of biomarkers to assess graft health. Furthermore, by mapping altered metabolites back to metabolic pathways it may also facilitate a better mechanistic understanding of the underlying pathology contributing to under or over-immunosuppression.

[0302] In this study we sought to identify altered metabolites correlated with overimmunosuppression. This could provide the so-called “missing guard-rail” in post-KT monitoring. We evaluated 108 urine samples from a total of 106 RTRs, 29 with biopsy proven PVAN and 79 with a stable graft and no sign of rejection or PVAN from a biopsy. PVAN is a marker of overimmunosuppression, and samples from RTRs with biopsy proven PVAN were used to represent the over-immunosuppression phenotype. To the best of our knowledge, this is one of the largest PVAN cohorts. Urine metabolites were first analyzed via the Olaris global unbiased NMR platform. This yielded a set of differential metabolites that formed the basis for a machine learning (ML) model with 91 % area under the curve (AUC), 89% accuracy, and 76% sensitivity and 92% specificity in training data; and 80% AUC, 80% accuracy, and 70% sensitivity and 85% specificity in a separate validation set to differentiate patients with PVAN from those with Stable grafts. In parallel, in a subset of patients (41 total samples from 38 RTRs including 12 PVAN samples and 29 Stable samples) we analyzedurine metabolites via a targeted LC-MS panel. In a similar manner, based on the differential metabolites, we constructed a ML algorithm with 85% AUC, 76% accuracy, and 75% sensitivity and 76% specificity to differentiate patients with PVAN from those with Stable grafts. Collectively these results suggest metabolites provide a promising class of biomarkers to identify overimmunosuppression and that both NMR and MS provide suitable platforms to develop a diagnostic assay to distinguish PVAN from Stable RTRs.METHODSNMR SAMPLE DESCRIPTION

[0303] Urine samples (N=108) from 106 RTR patients with matched pathology from biopsy were provided. This included 29 samples with biopsy proven PVAN (over-immunosuppressed) and 79 samples with a clean biopsy and no indication of rejection or infection (“Stable”) (Table 1). There were two RTRs that had serial urine samples and biopsies 13 and 29 days apart, both with PVAN. The mean age of the PVAN cohort was 58.27 years, while the stable patients, although not statistically significant, were slightly younger, with a mean age of 59.94 years. Both PVAN and Stable had more males than females but there was no significant difference in sex between groups. Days post-transplant, serum creatinine, and eGFR were comparable between groups. The was a slight increase in the percentage of indicative biopsies vs protocol biopsies for the PVAN group, but this was not statistically significant. Of note, 7 Stable samples were missing sex, 13 PVAN and 1 1 Stable samples lacked serum creatinine and 6 Stable samples were missing eGFR values in the clinical data.NMR SAMPLE PREPARATION

[0304] Proteins and macromolecules were removed, and metabolites extracted via methanol / chloroform precipitation. The aqueous layer was partially evaporated under reduced pressure and lyophilized overnight. Lyophilized samples were dissolved in 0.25 mL of sodium phosphate buffer (pH 7.4) in D2O. A total of 10 mM of deuterated sodium 2,2-dimethyl-2-silapentane- 5-sulfonate (D, 98% DSS-d6) was added to each sample for chemical shift referencing. A total of 200 pL of sample was then transferred to the NMR tube and analyzed.NMR DATA COLLECTION AND PROCESSING

[0305] Metabolites were analyzed via 1 D1H and 2D13C-1H HSQC NMR spectroscopy using a Bruker AVANCE II solution-state 600 MHz spectrometer equipped with a liquid helium-cooled Prodigy TCICryoprobe (H / F, C, N), using a noesyprld and hsqcetgpsisp2.2 pulse program and non-uniform sampling (NUS), respectively. 1 D spectra were processed on Topspin (Bruker Topspin 3.6.4, Bruker BioSpin, Rheinstetten, Germany), Matlab (Matlab R2015b, Mathworks Inc., Natick, MA), and SigMa S8. The acquired 2D spectral data were processed using the NMRPipe software package69. The NUS data were reconstructed using iterative soft thresholding according to the hmslST algorithm70. NMR data were zero-filled, Fourier-transformed and automatically phase-corrected to yield a final digital resolution of 2048 (N2) x 2048 (N1 ) points. The processed NMR data were then used to generate peak lists using NMRPipe. Metabolite resonances with a signal intensity below 7e-4 of total signal intensity per sample were excluded, the rest were dynamically binned into clusters using Density- Based Spatial Clustering of Applications with Noise (DBSCAN) in both dimensions and each cluster was given a unique XID number, e.g., X109, based on its chemical shift. Resonance clusters were normalized using Probabilistic Quotient Normalization (PQN) (REF) and filtered using a criterion of being present in at least 80% of all samples in one class. Chemical shift queries and metabolite annotations were performed using Olaris proprietary software.NMR QUALITY CONTROL

[0306] Proton (1H) NMR provides a powerful and robust quantitative comparison of metabolites (Bingol, 2018). However, chemical shift overlap limits the number of metabolites that can be accurately and unambiguously quantified. For this reason, the 1 D results (representative Stable and PVAN spectra in FIGS. 2A-2B) were used to observe overall trends and guide a more in-depth 2D 1 H-13C analysis, wherein by spreading the data across both1H and13C dimensions, wherein each resonance represents a carbon atom attached to a unique proton (a C-H pair). In this manner, one metabolite can be represented by multiple resonances (representative Stable and PVAN spectra in FIGS. 2C-2D). Samples were referenced using internal standard DSS. To assess any experimental or sample processing errors, we compared DSS intensity as well as the overall resonance intensity sum (total sum), mean, median, maximum, and number of peaks for each sample (FIG. 3A). Most of the samples had relatively consistent values, suggesting good quality data. The intensity of DSS was low in 4 samples (10443-S, 10483-S, 10490-S, 10492-S), which could be indicative of high protein and had to be discarded. An additional 4 samples (9233-P, 10432- S, 10458-S, 10404-P) had significantly higher total sum than the other samples. Visual inspection of these samples displayed “streaking” which could suggest degradation or other problems with the sample, and they were also removed. Principle component analysis (PGA) was calculated for the final sample set (N=100), using the singular value decomposition (SVD) method (FIG. 3B). PC1 , PC2 and PC3 accounted for 33.92%, 11 .24% and 9.35 % of the variance respectively. The majorityof samples clustered together suggesting a similar composition for statistical comparison. A few samples (9328-S, 10487-S, 9360-S, 9265-P and 10415-P) were flagged as potential outliers, however after manual inspection of each spectrum, no significant concerns were identified, and they were included in the analysis.LC-MS SAMPLE PREPARATION

[0307] For LC-MS analysis, urine samples were thawed on ice, briefly vortexed, and where necessary diluted with water to a specific gravity (SG) of 1.02, before transferring 25 pL of each sample to an Eppendorf tube. To extract urinary metabolites, 75 pL of an acetonitrile / methanol solution (1 :1 v / v) was added to each study sample, the mixture was vortexed, centrifuged at 4°C, and the supernatant carefully transferred to a LC-MS vial. Additionally, a pooled quality control (QC) sample was prepared by mixing equal aliquots of all study samples, as well as a blank sample consisting of water. The QC sample and the blank sample were extracted the same way as the study samples and all samples were stored at -20 °C until analysis.LC-MS SAMPLE ACQUISITION

[0308] Targeted mass spectrometric analysis of urinary metabolites was carried out using a Thermo Scientific Altis MD with a Thermo Scientific Vanquish Flex UHPLC system. All samples were analyzed in randomized order evenly interspersed by QC samples, by injecting 2 pL of the sample on a Waters Atlantis Premier BEH Z-HILIC VanGuard Fit column (2.1 mm x 150 mm, 1.7 pm). The flow rate was 350 pL / min with mobile phase A water / acetonitrile (95%: 5% v / v) containing 10 mM ammonium acetate, and mobile phase B water / acetonitrile (5%: 95% v / v) containingl O mM ammonium acetate. The mobile phase gradient program started at 5% A, held constant to 1 minutes, linearly increased to 25% A at 5 minutes, linearly increased to 40% A at 10 min, linearly increased to 55% A at 1 1.5 minute, held constant until 14.5 minutes and returned to starting conditions at 16 minutes and the acquisition stopped at 20.5 minutes. The column and autosampler temperatures were kept at 40 °C and 5 °C respectively. Ionization was performed using heated electrospray ionization (HESI) with a spray voltage of 4.5 kV for positive mode and 3.5 kV for negative mode. The vaporizer and the ion transfer tube temperatures were set at 350 °C and 325 °C respectively. A sheath gas flow of 40 (arbitrary units) and a sweep gas flow of 2 (arbitrary units) were applied for acquiring data. SRMs were previously experimentally determined using authentic standards and metabolite identity was confirmed using a combination of 2 SRMs for most metabolites as well as retention times obtained from the analysis of the authentic standards. Peak integration was performed with Skyline (REF).MS QUALITY CONTROL

[0309] To assess potential experimental or sample processing errors, we assessed the total sum, mean, median, and maximum feature intensities for each sample (FIG. 14A). Most samples displayed consistent values, indicating good data quality. Although sample 9546-P stood out as having lower values across all criteria, and sample 9473-S was low in median and maximum, this could have been due to low SG measurements (1 .007 for 9546-P and 1 .0038 for 9473-S), which can be resolved during with data normalization. As was done for the NMR data, we next performed PCA using SVD (FIG. 14B). PC1 , PC2, and PC3 accounted for 27.18%, 10.37%, and 8.01% of the variance, respectively. The majority of the study samples overlapped, indicating a similar composition for statistical comparison, and the pooled QC analyses clustered tightly together in the middle of all measured samples as expected . Although samples 9360-S and 9546-P were initially flagged as potential outliers, manual inspection of their spectra revealed no data quality concerns and were therefore also included in the downstream analysis.

[0310] To assess instrument performance, we evaluated the QC sample that was measured at regular intervals throughout the study (FIG. 14C). The coefficient of variation (CV) for each feature was computed, resulting in an average CV of 9.4% + / -10.5%, suggesting high quality and reproducible data throughout the experiment (FIG. 14D). Notably, three features (Flavin adenine dinucleotide [pos] - Tier 2, Dihydroxyacetone phosphate [neg] - Tier 2, and UDP-N-acetyl-alpha-D- glucosamine [pos]) exhibited CVs exceeding 50%. This high variation was driven by the fact the signals were extremely low.

[0311] Following the QC assessment, the LC-MS data underwent signal loss correction using the QC locally estimated scatterplot smoothing (QC-LOESS) method. Specifically, a LOESS model was established for each metabolite within the QC samples, with the signal as the dependent variable and the LC-MS sequence order as the independent variable. All samples were then normalized to this model to perform signal correction. After signal loss correction, adjusted intensity values below the established lower detection threshold (DLB) for the corresponding metabolite were set to 0. The DLB for each metabolite was set at 3x the mean intensity in the blank samples. The data were then normalized to each sample's median intensity and filtered based on the criterion of presence in at least 80% of samples within one class.Statistical Analysis and Machine Learning

[0312] Kruskal-Wallis (KW) non-parametric one-way analysis of variance (ANOVA) was used to test for significant differences between groups of interest from normalized data for both the NMR and MSseparately. The test of significance was determined by a p-value cutoff (p<0.05) and adjusted based on false discovery rate (FDR) for multiple hypothesis testing correction. Fold change (FC) was calculated as the ratio of the median intensities of the two groups. A FC cutoff of 1.5 was used to determine significant changes, indicating that metabolites with FC greater than 1 .5 or less than 0.67 were considered increased or decreased, respectively. All statistical analyses and ML modeling were performed using R 3.6.3 (REF, accessed 13 October 2023) with the following packages: tidyverse for data wrangling [REF], glmnet for linear regression model [REF], caret and Boruta for feature selection [44,70], randomForest and ropls for modeling [71 ,72], and ggplot2 and pROC for result visualization [73,74].RESULTS OVERVIEWDifferential Metabolites Comparing PVAN vs Stable (NMR)

[0313] As a first step to identify metabolites correlated with PVAN, a bootstrap Kruskal Wallis (KW) test of significance was performed. A subset of the data (70%) was randomly selected, and a KW test of significance was performed comparing metabolite resonances in PVAN samples compared to those in Stable samples. This was repeated 1000 times. Significant features were identified as those with a p-value <0.05, fold-change (FC) >1 .5, and coefficient of variation (CV) <100% per class. In this manner we were able to rank the frequency each metabolite feature was found significantly different (FIG. 4A). For example, 2 features (X42 and X98) were only significantly different in 1 the KW splits, while 3 features (X109, X205 and X40) were significant in all 1000 tests of significance. There were 6 features (X109, X205, X40, X44, X1103, and X32) that were consistently altered in >70% of all KW tests. From the violin plots (FIG. 4B) we observed that each of these metabolite resonances was increased in PVAN compared to Stable.Building a PVAN vs Stable Classifier (NMR)

[0314] To build a machine learning (ML) classifier, the data was split into training (70%) and validation set (30%), wherein the ratio of PVAN to Stable, age, sex, and % indicative biopsy was consistent between both groups (FIGS. 5A-5B). The training data has a total of 70 samples including 17 PVAN samples and 53 Stable. The validation data (N=30), which was set aside for evaluation of the model, had 10 PVAN and 20 Stable samples.

[0315] Using the training samples, pre-defined feature sets were evaluated in distinct crossvalidated ML algorithms to classify PVAN from Stable and assessed based on accuracy, area under the curve (AUG), sensitivity and specificity. The pre-defined feature sets included the top 6 differentialfeatures (X109, X205, X40, X44, X1103, and X32) that passed the KW test and 5 features refined by the Boruta feature refinement model (X109, X205, X40, X44, and X32), which selects features based on the Random Forest (RF) importance of the unchanged vs shuffled feature (Kursa et al, 2010).

[0316] Using both feature sets, the training data was then split 70 / 30 into training and test sets, and 10-fold cross validated ML models were constructed using elastic net (ENET), orthogonal projections to latent structures discriminant analysis (OPLS-DA) and random forest (RF) models, leading to 6 ML models (FIG. 6). Model selection started with comparing the performance and stability of the cross-validated ML models between the training and test splits. Pre-evaluation of the models excluded 3 unstable, and potentially overfitting, models (FIG. 6, red outlines) because at least one of the four performance metrics was significantly different between the training and test datasets (p < 0.05, Student’s t-test). For the remaining 3 models, the average cross-validated accuracy, AUG, sensitivity, and specificity were 83.9 ± 3.2%, 85.6 ± 6.6%, 53.3 ± 8.7%, 93.5 ± 1.5%, suggesting highly consistent results across models. Based on the highest AUC of the individual cross-validated models, the OPLS model using all 6 features (FIG. 6, green outline) was selected as the top model.

[0317] The OPLS rankings were then used to derive a proprietary myOLARIS™ PVAN score that ranges from 0 to 1. The optimal cutoff score was fine tuned to maximize both sensitivity and specificity, with PVAN samples scoring over 0.28 and Stable samples under 0.28. Applying the myOLARIS™ PVAN Score to the full training data had accuracy 81.4%, AUC 90.2%, sensitivity 88.2%, and specificity 79.2%. This is visualized using a waterfall plot for the training dataset (FIG. 7, left). For perfect performance, PVAN patients (shown in red) would all be above the cutoff, while Stable patients (green) would all be below the cutoff. In the waterfall plots for the top model, the performance was excellent, with most PVAN patients above 0.28 and most Stable patients below 0.28. The model was then applied to the validation set of 30 samples including 10 PVAN samples and 20 stable samples which also had high performance with 73.3% accuracy, 79.5% AUC, 80.0% sensitivity, and 70.0% specificity, and is visualized by a waterfall plot (FIG. 7, right).

[0318] We next sought to annotate the 6 features (X109, X205, X40, X44, X1103, and X32) by comparing their chemical shifts to reference libraries of known metabolites and through spike-in experiments with standards (FIG. 8). We were able to confirm X109 corresponds to lysine (FIG. 8B). Upon addition of authentic lysine to the samples we saw an increase in intensity of X109 for both the HSQC and HSQC-TOCSY which provides cross peaks from the same spin system, enabling us to determine which resonances are from the same metabolite. Lysine is an essential proteogenic amino acid, important for protein synthesis, crosslinking of collagen peptides, uptake of nutrients, and the production of carnitine (REF). Altered lysine metabolism in the kidney has been implicated in diseaseprocesses (REF). Further, high doses of lysine have been shown to induce acute renal failure in animal models. The observed increase of lysine in PVAN samples could be reflective of the kidney damage associated with PVAN nephritis. X205, X40, X44 and X32, while clear and robust peaks (FIG. 8C), did not match with confidence to any known compounds in our library and efforts are underway to determine the metabolite identity. This could suggest novel biomarkers are altered in PVAN RTRs. X1103 putatively matched to several sugar metabolites, and the ability to discriminate between sugars is significantly challenging. From the Boruta feature selection, it was previously noted that X1 103 could be removed with only a slight decrease in model performance. Thus, we removed X1103, and refined the myOLARIS PVAN Score. In the training data this led to an accuracy of 80.0%, AUG 90.1 %, sensitivity of 88.2% and specificity of 77.4%, and in the validation data of 73.3% accuracy, 78.8% AUG, sensitivity of 80.0% and specificity of 70.0% (FIG. 9A). Overall, we maintained excellent ability to differentiate PVAN from Stable. Removing any one of the others feature led to significantly decreased AUC and sensitivity (FIG. 9B), demonstrating that all remaining 5 features are required for the model performance.

[0319] The myOLARIS-PVAN Score could be launched as an NMR-diagnostic as is. However, to port the assay to a MS, a defined set of differential metabolites and normalization metabolites are required. The current PQN technique derives a normalization factor per sample by calculating the median ratio of metabolite resonance intensities to a reference spectrum which in this case includes 88 features. We sought to identify a smaller set of normalization features to mimic the PQN. To accomplish this, we created a linear regression model with least absolute shrinkage and selection operator (LASSO) to predict the PQN factors using defined features. As a first step, highly correlated features (>0.97) were removed, reducing total feature set to 68. This set of features as predictors were then fitted to a cross-validated LASSO model with the PQN factors as the response variables. A selection of the top 10 features (X205, X108, X48, X11 , X280, X120, X59, X63, X83 and X92) was achieved with regularization parameter lambda of 0.03. Next, the LASSO model was retrained based on the 10 features. Finally, the predicted normalization factors using LASSO model were plotted against the PQN factors, with a correlation of 0.77 (FIG. 10). The high correlation suggests the LASSO model with these 10 normalization features (NFs) could substitute for the PQN factors.

[0320] To build our champion model, we rebuilt the myOLARIS-PVAN Score using the top 5 differential features with the 10 NFs. In the training data this led to an accuracy of 88.6%, AUC 91 .1%, sensitivity of 76.5% and specificity of 92.5%, and in the validation data of 80.0% accuracy, 79.5% AUC, sensitivity of 70.0% and specificity of 85.0% (FIG. 11 A). We also performed a stress test of each of the 5 differential features that signaled each feature is important for modelperformance (FIG. 11 B). The champion model with defined features has excellent ability to differentiate PVAN from Stable.

[0321] As described previously, we then sought to annotate the 10 NFs by comparing their chemical shift to known libraries and through spike in experiments. Of the 10 NFs we were able to confirm the identify of 6 including X13, X120, X59, X63, X83 and X92, which correspond to hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide (TMAO) and glutamine respectively (FIG. 12). X205, X108, X48 did not match any known compound in our library. X280 falls in the spectra indicative of a fatty acid. Efforts are underway to further resolve the identify of all unknown metabolites in the champion model.

[0322] The high accuracy of the myOLARIS-PVAN Score accomplished our goal of this evaluation study, wherein we were successfully able to, “Identify urine-based metabolite signature of biopsy proven PVAN compared to stable RTRs with pathology free of rejection or infection.” This is an exciting ste...

Claims

What is claimed is:

1. A method of monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy, the method comprising:(a) obtaining a urine sample from the renal transplant recipient; and(b) measuring levels of one or more metabolites in the urine sample using nuclear magnetic resonance (NMR), wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, intensities of one or more differential NMR metabolite features of one or more metabolites in the urine sample are used to classify the kidney graft as stable or unstable at risk of injury using a first machine learning model, wherein the one or more differential NMR metabolite features are selected from X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.67 ’H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.17 ’H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51 .0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm; and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, clinical information comprising gender of the renal transplant recipient, age of the renal transplant recipient at time of transplant of the kidney graft, plasma level of BK polyomavirus (BKV) in the renal transplant recipient, level of serum creatinine in the renal transplant recipient, and estimated glomerular filtration rate (eGFR) of the renal transplant recipient in combination with intensities of one or more differential NMR metabolite features of one or more metabolites in the urine sample are used to classify the kidney graft as stable or unstable at risk of injury using a second machine learning model, wherein the one or more differential NMR metabolite features are selected from an NMR metabolite feature having a chemical shift at 4.261H ppm and a chemical shift at 68.9913C ppm, an NMR metabolite feature having a chemical shift at 3.871H ppm and a chemical shift at 58.813C ppm, an NMR metabolite feature having a chemical shift at 7.141H ppm and a chemical shift at 119.9313C ppm, an NMR metabolite feature having a chemical shift at 3.221H ppm and a chemical shift at 30.1213C ppm, an NMR metabolite feature having a chemical shift at 8.181H ppm and a chemical shift at 137.9213C ppm, and one or more NMR metabolite features for one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoicacid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine.

2. The method of claim 1 , wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, the one or more differential NMR metabolite features comprise the X65 having the chemical shift at 4.061H ppm and the chemical shift at 85.9413C ppm, the X56 having the chemical shift at 4.671H ppm and the chemical shift at 98.5713C ppm, the X360 having the chemical shift at 2.171H ppm and the chemical shift at 24.5113C ppm, the X333 having the chemical shift at 4.441H ppm and the chemical shift at 51.0413C ppm, the X93 having the chemical shift at 4.081H ppm and the chemical shift at 74.8313C ppm, the X129 having the chemical shift at 3.281H ppm and the chemical shift at 62.2313C ppm, and the X161 having the chemical shift at 3.971H ppm and the chemical shift at 46.5513C ppm; and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, the one or more differential NMR metabolite features comprise the NMR metabolite feature having the chemical shift at 4.261H ppm and the chemical shift at 68.9913C ppm, the NMR metabolite feature having the chemical shift at 3.871H ppm and the chemical shift at 58.813C ppm, the NMR metabolite feature having the chemical shift at 7.141H ppm and a chemical shift at 119.9313C ppm, the NMR metabolite feature having the chemical shift at 3.221H ppm and the chemical shift at 30.1213C ppm, the NMR metabolite feature having the chemical shift at 8.181H ppm and the chemical shift at 137.9213C ppm, and the one or more NMR metabolite features for the one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5- dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine.

3. The method of claim 1 or 2, wherein if the kidney graft is classified as unstable at risk of injury, the method further comprises determining whether the renal transplant recipient is underimmunosuppressed or over-immunosuppressed, wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, clinical information comprising the age of the renal transplant recipient at the time of transplant of the kidney graft and the plasmalevel of BKV in the renal transplant recipient in combination with intensities of differential NMR metabolite features comprising X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm and X2473 having a chemical shift at 3.811H ppm and a chemical shift at 63.9313C ppm are used to classify the renal transplant recipient as under-immunosuppressed or overimmunosuppressed using a third machine learning model, and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, clinical information comprising the gender of the renal transplant recipient, the age of the renal transplant recipient at time of transplant of the kidney graft, the plasma level of BKV in the renal transplant recipient, the level of serum creatinine in the renal transplant recipient, and the eGFR of the renal transplant recipient in combination with intensities of the one or more differential NMR metabolite features selected from the NMR metabolite feature having the chemical shift at 4.261H ppm and the chemical shift at 68.9913C ppm, the NMR metabolite feature having the chemical shift at 3.871H ppm and the chemical shift at 58.813C ppm, the NMR metabolite feature having the chemical shift at 7.141H ppm and the chemical shift at 1 19.9313C ppm, the NMR metabolite feature having the chemical shift at 3.221H ppm and the chemical shift at 30.1213C ppm, the NMR metabolite feature having the chemical shift at 8.181H ppm and the chemical shift at 137.9213C ppm, and the one or more NMR metabolite features for the one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine are used to classify the renal transplant recipient as under-immunosuppressed or overimmunosuppressed using a fourth machine learning model.

4. The method of claim 3, further comprising altering the immunosuppressive therapy administered to the renal transplant recipient if the renal transplant recipient is determined to be under-immunosuppressed or over-immunosuppressed, wherein immunosuppression is increased if the renal transplant recipient is determined to be under-immunosuppressed and decreased if the renal transplant recipient is determined to be over-immunosuppressed.

5. The method of claim 3 or 4, further comprising performing diagnostic screening to determine whether the renal transplant recipient has an infection, a malignancy, or polyomavirus- associated nephropathy if the renal transplant recipient is determined to be overimmunosuppressed; and treating the renal transplant recipient for the infection, the malignancy, orthe polyomavirus-associated nephropathy if the renal transplant recipient is diagnosed as having the infection, the malignancy, or the polyomavirus-associated nephropathy.

6. The method of claim 3 or 4, further comprising performing diagnostic screening to determine whether the renal transplant recipient has graft dysfunction, subclinical graft rejection, or graft rejection if the renal transplant recipient is determined to be under-immunosuppressed; and treating the renal transplant recipient for the graft dysfunction, subclinical graft rejection, or graft rejection if the renal transplant recipient is diagnosed as having the graft dysfunction, subclinical graft rejection, or graft rejection.

7. The method of any one of claims 1 -6, wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, the one or metabolites comprise glucuronic acid, trigonelline, trimethylamine N-oxide (TMAO), and hippuric acid.

8. The method of any one of claims 1 -7, wherein the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA) or a random forest (RF) machine learning algorithm.

9. The method of any one of claims 1 -8, wherein the NMR spectroscopy comprises onedimensional (1 D)1H NMR spectroscopy, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectroscopy, or a combination thereof.

10. The method of claim 9, wherein the NMR is performed with non-uniformed sampling (NUS).

11. The method of any one of claims 1 -10, further comprising using one or more normalization NMR metabolite features.

12. A computer implemented method for monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy, the computer performing steps comprising:(a) receiving nuclear magnetic resonance (NMR) spectra of one or more metabolites from a urine sample obtained from the renal transplant recipient;(b) measuring intensities of one or more differential NMR metabolite features in the NMR spectra, wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, the intensities of the one or more differential NMR metabolite features of the one or more metabolites in the urine sample are used to classify the kidney graft as stable or unstable at risk of injury using a first machine learning model, wherein the one or more differential NMR metabolite features are selected from X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.94 13C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm; and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, clinical information comprising gender of the renal transplant recipient, age of the renal transplant recipient at time of transplant of the kidney graft, plasma level of BK polyomavirus (BKV) in the renal transplant recipient, level of serum creatinine in the renal transplant recipient, and estimated glomerular filtration rate (eGFR) of the renal transplant recipient in combination with the intensities of one or more differential NMR metabolite features of the one or more metabolites in the urine sample are used to classify the kidney graft as stable or unstable at risk of injury using a second machine learning model, wherein the one or more differential NMR metabolite features are selected from an NMR metabolite feature having a chemical shift at 4.261H ppm and a chemical shift at 68.9913C ppm, an NMR metabolite feature having a chemical shift at 3.871H ppm and a chemical shift at 58.813C ppm, an NMR metabolite feature having a chemical shift at 7.141H ppm and a chemical shift at 119.9313C ppm, an NMR metabolite feature having a chemical shift at 3.221H ppm and a chemical shift at 30.1213C ppm, an NMR metabolite feature having a chemical shift at 8.181H ppm and a chemical shift at 137.9213C ppm, and one or more NMR metabolite features for one or more metabolites selected from 4-hydroxy-3- methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine; and(c) displaying information regarding whether the kidney graft is stable or unstable at risk of injury.

13. The computer implemented method of claim 12, wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, the one or more differential NMR metabolite features comprise the X65 having the chemical shift at 4.061H ppm and the chemical shift at 85.9413C ppm, the X56 having the chemical shift at 4.67 ’H ppm and the chemical shift at 98.5713C ppm, the X360 having the chemical shift at 2.171H ppm and the chemical shift at 24.5113C ppm, the X333 having the chemical shift at 4.441H ppm and the chemical shift at 51.0413C ppm, the X93 having the chemical shift at 4.081H ppm and the chemical shift at 74.8313C ppm, the X129 having the chemical shift at 3.281H ppm and the chemical shift at 62.2313C ppm, and the X161 having the chemical shift at 3.971H ppm and the chemical shift at 46.5513C ppm; and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, the one or more differential NMR metabolite features comprise the NMR metabolite feature having the chemical shift at 4.261H ppm and the chemical shift at 68.9913C ppm, the NMR metabolite feature having the chemical shift at 3.871H ppm and the chemical shift at 58.813C ppm, the NMR metabolite feature having the chemical shift at 7.141H ppm and a chemical shift at 119.9313C ppm, the NMR metabolite feature havingthe chemical shift at 3.221H ppm and the chemical shift at 30.1213C ppm, the NMR metabolite feature having the chemical shift at 8.181H ppm and the chemical shift at 137.9213C ppm, and the one or more NMR metabolite features for the one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5- dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine.

14. The computer implemented method of claim 12 or 13, wherein if the kidney graft is classified as unstable at risk of injury, the method further comprises determining whether the renal transplant recipient is under-immunosuppressed or over-immunosuppressed, wherein if the urine sample is obtained from the renal transplant recipient within 180 days after transplant of the kidney graft into the renal transplant recipient, clinical information comprising the age of the renal transplant recipient at the time of transplant of the kidney graft and the plasma level of BKV in the renal transplant recipient in combination with the intensities of differential NMRmetabolite features comprising X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm and X2473 having a chemical shift at 3.811H ppm and a chemical shift at 63.9313C ppm are used to classify the renal transplant recipient as under-immunosuppressed or overimmunosuppressed using a third machine learning model, and wherein if the urine sample is obtained at 180 days or more after transplant of the kidney graft into the renal transplant recipient, clinical information comprising the gender of the renal transplant recipient, the age of the renal transplant recipient at time of transplant of the kidney graft, the plasma level of BKV in the renal transplant recipient, the level of serum creatinine in the renal transplant recipient, and the eGFR of the renal transplant recipient in combination with the intensities of the one or more differential NMR metabolite features selected from the NMR metabolite feature having the chemical shift at 4.261H ppm and the chemical shift at 68.9913C ppm, the NMR metabolite feature having the chemical shift at 3.871H ppm and the chemical shift at 58.813C ppm, the NMR metabolite feature having the chemical shift at 7.141H ppm and the chemical shift at 1 19.9313C ppm, the NMR metabolite feature having the chemical shift at 3.221H ppm and the chemical shift at 30.1213C ppm, the NMR metabolite feature having the chemical shift at 8.181H ppm and the chemical shift at 137.9213C ppm, and the one or more NMR metabolite features for the one or more metabolites selected from 4-hydroxy-3-methylbenzoic acid, homogenetistic acid, lysine, quinic acid, threonine, pantothenic acid, caffeine, asparagine, 2,5-dihydroxybenzoic acid, ketoleucine, citrulline, 2,3-dihydroxybenzoic acid, pyroglutamic acid, tyramine, mycophenolic acid acyl-glucuronide (acMPAG), cystine, arginine, glutamine, hippuric acid, alanine, phosphorylcholine, glycine, and taurine are used to classify the renal transplant recipient as under-immunosuppressed or overimmunosuppressed using a fourth machine learning model; and displaying information regarding whether the renal transplant recipient is underimmunosuppressed or over-immunosuppressed.

15. The computer implemented method of any one of claims 12-14, further comprising storing the information regarding whether the renal transplant recipient is under-immunosuppressed or over-immunosuppressed in a database.

16. The computer implemented method of any one of claims 12-15, wherein the NMR spectra comprise one-dimensional (1 D)1H NMR spectra, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectra, or a combination thereof.

17. The computer implemented method of claim 16, wherein the NMR spectra are obtained using non-uniformed sampling (NUS).

18. A system for monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy using the computer implemented method of any one of claims 12-17, the system comprising:(a) a storage component for storing data, wherein the storage component has instructions for monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy based on analysis of the NMR spectra of the one or more metabolites stored therein;(b) a computer processor programmed to analyze the NMR spectra using one or more algorithms, wherein the computer processor is coupled to the storage component and configured to execute the instructions stored in the storage component in order to receive the inputted NMR spectra, and analyze the NMR spectra according to the computer implemented method of any one of claims 12-17; and(c) a display component for displaying information regarding whether the renal transplant recipient is stable, unstable at risk of injury, under-immunosuppressed, or over-immunosuppressed.

19. A non-transitory computer-readable medium comprising program instructions that, when executed by a processor in a computer, causes the processor to perform the computer implemented method of any one of claims 12-17.

20. A kit comprising the non-transitory computer-readable medium of claim 19 and instructions for monitoring a kidney graft in a renal transplant recipient undergoing treatment with immunosuppressive therapy.

21. A method of detecting over-immunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy, the method comprising:(a) obtaining a urine sample from the renal transplant recipient;(b) measuring levels of one or more metabolites in the urine; and(c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed.

22. The method of claim 21 , further comprising altering the immunosuppressive therapy administered to the renal transplant recipient if the renal transplant recipient is determined to be over-immunosuppressed.

23. The method of claim 21 or 22, further comprising performing diagnostic screening to determine whether the renal transplant recipient has an infection, a malignancy, or polyomavirus- associated nephropathy if the renal transplant recipient is determined to be overimmunosuppressed; and treating the renal transplant recipient for the infection, the malignancy, or the polyomavirus-associated nephropathy if the renal transplant recipient is diagnosed as having the infection, the malignancy, or the polyomavirus-associated nephropathy.

24. The method of claim 23, wherein the infection is a BK virus infection.

25. The method of any one of claims 21 -24, wherein said measuring comprises performing nuclear magnetic resonance (NMR) spectroscopy or mass spectrometry.

26. The method of claim 25, wherein the NMR spectroscopy is multi-dimensional NMR spectroscopy.

27. The method of claim 25, wherein the NMR spectroscopy comprises one-dimensional (1 D)1H NMR spectroscopy, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectroscopy, or a combination thereof.

28. The method of claim 26 or 27, wherein the NMR is performed with non-uniformed sampling (NUS).

29. The method of any one of claims 25-28, wherein one or more differential NMR metabolite features are used to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model, wherein the one or more differential NMR metabolite features are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061Hppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.

30. The method of claim 29, wherein the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.31 . The method of claim 29, wherein the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.

32. The method of any one of claims 25-31 , further comprising using one or more normalization NMR metabolite features in combination with the differential NMR metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

33. The method of claim 32, wherein the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

34. The method of claim 33, wherein the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1 .561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

35. The method of claim 32, wherein the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.

36. The method of any one of claims 25-35, wherein chemical shifts are calibrated against an internal deuterated sodium 2,2-dimethyl-2-silapentane-5-sulfonate standard.

37. The method of any one of claims 21 -36, wherein said measuring comprises performing liquid chromatography-mass spectrometry (LC-MS).

38. The method of claim 37, wherein ionization is performed using heated electrospray ionization (HESI).

39. The method of claim 37 or 38, wherein the liquid chromatography is performed using a linear gradient of 5% to 95% acetonitrile.

40. The method of any one of claims 37-39, wherein one or more differential LC-MS metabolite features of caffeine, benzoyl formic acid, and cytidine are used to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.41 . The method of claim 40, wherein the differential LC-MS metabolite features for caffeine comprise a feature with a mass-to-charge ratio (m / z) of 195.00 for a caffeine ([M+H]+ ionand a feature with a m / z of 137.90 and a feature with a m / z of 110.00 from selected reaction monitoring.

42. The method of claim 40 or 41 , wherein the differential LC-MS metabolite features for benzoyl formic acid comprise a feature with a m / z of 149.09 for a benzoyl formic acid [M-H]- ion and a feature with a m / z of 104.91 and a feature with a m / z of 77.07 from selected reaction monitoring.

43. The method of any one of claims 40-42, wherein the differential LC-MS metabolite features for cytidine comprise a feature with a m / z of 244.09 for a cytidine [M+H]+ ion and a feature with a m / z of 111 .90 from selected reaction monitoring.

44. The method of any one of claims 40-43, further comprising using one or more normalization LC-MS metabolite features in combination with the differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

45. The method of claim 44, wherein the normalization LC-MS metabolite features comprise lysine, hippuric acid, mannitol, trimethylamine N-oxide, and glutamine LC-MS features.

46. The method of any one of claims 40-45, wherein one or more differential NMR metabolite features are used in combination with one or more differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model, wherein the one or more differential NMR metabolite features are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm; and wherein the one or more differential LC-MS metabolite features are selected from a feature with a mass-to-charge ratio (m / z) of 195.00 for a caffeine ([M+H]+ ion and a feature with a m / z of 137.90 and a feature with a m / z of 110.00 from selected reaction monitoring, a feature with a m / z of 149.09 for a benzoyl formic acid [M-H]- ion and a feature with a m / z of 104.91 and a feature with am / z of 77.07 from selected reaction monitoring, a feature with a m / z of 244.09 for a cytidine [M+H]+ ion and a feature with a m / z of 111 .90 from selected reaction monitoring.

47. The method of claim 46, wherein the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.

48. The method of claim 46, wherein the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.

49. The method of any one of claims 46-48, further comprising using one or more normalization NMR metabolite features and normalization LC-MS metabolite features in combination with the differential NMR metabolite features and differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

50. The method of claim 49, wherein the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.56 ’H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.51 . The method of claim 50, wherein the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1 .561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

52. The method of claim 49, wherein the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.

53. The method of any one of claims 49-52, wherein the normalization LC-MS metabolite features comprise lysine, hippuric acid, mannitol, trimethylamine N-oxide, and glutamine LC-MS features.

54. The method of any one of claims 21-53, further comprising isolating metabolites from the urine sample prior to said measuring.

55. The method of any one of claims 21 -54, wherein the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA) or a random forest (RF) machine learning algorithm.

56. The method of any one of claims 21 -55, further comprising repeating steps (a) - (c) periodically to determine whether the renal transplant recipient has become overimmunosuppressed.

57. The method of claim 56, wherein steps (a) - (c) are repeated at least twice a month, at least once a month, at least every 2 months, at least every 3 months, at least every 4 months, at least every 5 months, at least every 6 months, or at least once a year.

58. The method of any one of claims 21-57, wherein the one or more metabolites are selected from lysine, caffeine, benzoyl formic acid, and cytidine.

59. A biomarker selected from lysine, caffeine, benzoyl formic acid, and cytidine for use in a method of diagnosing over-immunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy.

60. A kit comprising: a container for collecting a urine sample from a renal transplant recipient, and instructions to halt, alter, or monitor treatment of the renal transplant recipient based on analyzing the levels of one or more metabolites in the urine sample using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed according to the method of any one of claims 21 -58.61 . A computer implemented method for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed, the computer performing steps comprising:(a) receiving nuclear magnetic resonance (NMR) and / or liquid chromatography-mass spectrometry (LC-MS) spectra of one or more metabolites from a urine sample obtained from the renal transplant recipient;(b) measuring levels of the one or more metabolites using the NMR and / or LC-MS spectra;(c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed; and(d) displaying information regarding whether the renal transplant recipient is overimmunosuppressed.

62. The computer implemented method of claim 61 , further comprising storing the information regarding whether the renal transplant recipient is over-immunosuppressed in a database.

63. The computer implemented method of claim 61 or 62, wherein the NMR spectra are multi-dimensional NMR spectra.

64. The computer implemented method of claim 63, wherein the NMR spectra comprise one-dimensional (1 D)1H NMR spectra, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectra, or a combination thereof.

65. The computer implemented method of claim 63 or 64, wherein the NMR spectra are obtained using non-uniformed sampling (NUS).

66. The computer implemented method of any one of claims 61-65, wherein one or more differential NMR metabolite features in the NMR spectra are used to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model, wherein the one or more differential NMR metabolite features are selected from X109 having a chemical shift at 3.01 ’H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.

67. The computer implemented method of claim 66, wherein the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.

68. The computer implemented method of claim 66, wherein the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.

69. The computer implemented method of any one of claims 61 -68, further comprising using one or more normalization NMR metabolite features in combination with the differential NMRmetabolite features in the NMR spectra to classify the renal transplant recipient as over- immunosuppressed or stable using the machine learning model.

70. The computer implemented method of claim 69, wherein the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1 .561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

71. The computer implemented method of claim 70, wherein the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

72. The computer implemented method of claim 69, wherein the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.

73. The computer implemented method of any one of claims 61 -72, wherein chemical shifts are calibrated against an internal deuterated sodium 2,2-dimethyl-2-silapentane-5-sulfonate standard.

74. The computer implemented method of any one of claims 61-73, wherein one or more differential LC-MS metabolite features of caffeine, benzoyl formic acid, and cytidine in the LC-MS spectra are used to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

75. The computer implemented method of claim 74, wherein the differential LC-MS metabolite features for caffeine comprise a feature with a mass-to-charge ratio (m / z) of 195.00 for a caffeine ([M+H]+ ion and a feature with a m / z of 137.90 and a feature with a m / z of 110.00 from selected reaction monitoring.

76. The computer implemented method of claim 74 or 75, wherein the differential LC-MS metabolite features for benzoyl formic acid comprise a feature with a m / z of 149.09 for a benzoyl formic acid [M-H]- ion and a feature with a m / z of 104.91 and a feature with a m / z of 77.07 from selected reaction monitoring.

77. The computer implemented method of any one of claims 74-76, wherein the differential LC-MS metabolite features for cytidine comprise a feature with a m / z of 244.09 for a cytidine [M+H]+ ion and a feature with a m / z of 111 .90 from selected reaction monitoring.

78. The computer implemented method of any one of claims 74-77, further comprising using one or more normalization LC-MS metabolite features in combination with the differential LC- MS metabolite features in the LC-MS spectra to classify the renal transplant recipient as overimmunosuppressed or stable using the machine learning model.

79. The computer implemented method of claim 78, wherein the normalization LC-MS metabolite features comprise lysine, hippuric acid, mannitol, trimethylamine N-oxide, and glutamine LC-MS features.

80. The computer implemented method of any one of claims 61-79, wherein one or more differential NMR metabolite features are used in combination with one or more differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model, wherein the one or more differential NMR metabolite features are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shiftat 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.97 ’H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm; and wherein the one or more differential LC-MS metabolite features are selected from a feature with a mass-to-charge ratio (m / z) of 195.00 for a caffeine ([M+H]+ ion and a feature with a m / z of 137.90 and a feature with a m / z of 110.00 from selected reaction monitoring, a feature with a m / z of 149.09 for a benzoyl formic acid [M-H]- ion and a feature with a m / z of 104.91 and a feature with a m / z of 77.07 from selected reaction monitoring, a feature with a m / z of 244.09 for a cytidine [M+H]+ ion and a feature with a m / z of 111 .90 from selected reaction monitoring.81 . The computer implemented method of claim 80, wherein the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.

82. The computer implemented method of claim 80, wherein the differential NMR metabolite features comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.

83. The computer implemented method of any one of claims 80-82, further comprising using one or more normalization NMR metabolite features and normalization LC-MS metabolite features in combination with the differential NMR metabolite features and differential LC-MS metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

84. The computer implemented method of claim 83, wherein the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1 .561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

85. The computer implemented method of claim 84, wherein the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.24 ’H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

86. The computer implemented method of claim 83, wherein the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.

87. The computer implemented method of any one of claims 83-86, wherein the normalization LC-MS metabolite features comprise lysine, hippuric acid, mannitol, trimethylamine N- oxide, and glutamine LC-MS features.

88. The computer implemented method of any one of claims 61 -87, wherein the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA) or a random forest (RF) machine learning algorithm.

89. A system for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed using the computer implemented method of any one of claims 61 -88, the system comprising:(a) a storage component for storing data, wherein the storage component has instructions for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed based on analysis of the NMR and / or LC-MS spectra of the one or more metabolites stored therein;(b) a computer processor programmed to analyze the NMR and / or LC-MS spectra using one or more algorithms, wherein the computer processor is coupled to the storage component and configured to execute the instructions stored in the storage component in order to receive the inputted NMR and / or LC-MS spectra, and analyze the NMR and / or LC-MS spectra according to the computer implemented method of any one of claims 61 -88; and(c) a display component for displaying information regarding whether the renal transplant recipient is over-immunosuppressed.

90. A non-transitory computer-readable medium comprising program instructions that, when executed by a processor in a computer, causes the processor to perform the computer implemented method of any one of claims 61 -88.

91. A kit comprising the non-transitory computer-readable medium of claim 90 and instructions for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed.

92. A method of detecting over-immunosuppression or under-immunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy, the method comprising:(a) obtaining a urine sample from the renal transplant recipient;(b) measuring levels of one or more metabolites in the urine; and(c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed or underimmunosuppressed.

93. The method of claim 92, wherein the one or more metabolites are selected from lysine, caffeine, benzoyl formic acid, cytidine, glucuronic acid, glucaric acid, lactose, glucose-6- phosphate, 1 -methylhistamine, histidine, trigonelline, trimethylamine N-oxide (TMAO), hippuric acid, choline, phosphoethanolamine, phosphocholine, acetylglycine, phenyllactic acid, mycophenolic acid glucuronide (MPAG), creatine, and hydroxyphenylacetic acid.

94. The method of claim 92 or 93, further comprising altering the immunosuppressive therapy administered to the renal transplant recipient if the renal transplant recipient is determined to be over-immunosuppressed or under-immunosuppressed, wherein immunosuppression is decreased if the renal transplant recipient is determined to be over-immunosuppressed and increased if the renal transplant recipient is determined to be under-immunosuppressed.

95. The method of any one of claims 92-94, further comprising performing diagnostic screening to determine whether the renal transplant recipient has an infection, a malignancy, or polyomavirus-associated nephropathy if the renal transplant recipient is determined to be overimmunosuppressed; and treating the renal transplant recipient for the infection, the malignancy, or the polyomavirus-associated nephropathy if the renal transplant recipient is diagnosed as having the infection, the malignancy, or the polyomavirus-associated nephropathy.

96. The method of any one of claims 92-94, further comprising performing diagnostic screening to determine whether the renal transplant recipient has graft dysfunction, subclinical graft rejection, or graft rejection if the renal transplant recipient is determined to be underimmunosuppressed; and treating the renal transplant recipient for the graft dysfunction, subclinical graft rejection, or graft rejection.

97. The method of any one of claims 92-96, wherein the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA) or a random forest (RF) machine learning algorithm.

98. The method of any one of claims 92-97, wherein said measuring comprises performing nuclear magnetic resonance (NMR) spectroscopy or mass spectrometry.

99. The method of claim 98, wherein the NMR spectroscopy is multi-dimensional NMR spectroscopy.

100. The method of claim 99, wherein the NMR spectroscopy comprises one-dimensional (1 D)1H NMR spectroscopy, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectroscopy, or a combination thereof.

101. The method of claim 99 or 100, wherein the NMR is performed with non-uniformed sampling (NUS).

102. The method of any one of claims 98-101 , wherein one or more differential NMR metabolite features are used to classify the renal transplant recipient as over-immunosuppressed, under-immunosuppressed, or stable using the machine learning model, wherein the one or more differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm; and wherein the one or more differential NMR metabolite features used to classify the renal transplant recipient as under-immunosuppressed or stable are selected from X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X1223 having a chemical shift at 7.941H ppm and a chemical shift at 140.3013C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.

103. The method of claim 102, wherein the differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable comprise or consist of X109 having a chemical shift at 3.01 ’H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shiftat 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.

104. The method of claim 102, wherein the differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable comprise or consist of X109 having a chemical shift at 3.01 ’H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.

105. The method of any one of claims 98-104, wherein the differential NMR metabolite features used to classify the renal transplant recipient as under-immunosuppressed or stable comprise or consist of X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.

106. The method of any one of claims 98-105, further comprising using one or more normalization NMR metabolite features in combination with the differential NMR metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

107. The method of claim 106, wherein the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1.56 ’H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

108. The method of claim 107, wherein the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1 .561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

109. The method of claim 106, wherein the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N- oxide, and glutamine NMR features.

110. The method of any one of claims 98-109, wherein chemical shifts are calibrated against an internal deuterated sodium 2,2-dimethyl-2-silapentane-5-sulfonate standard.

111. The method of any one of claims 92-1 10, further comprising repeating steps (a) - (c) periodically to determine whether the renal transplant recipient has become overimmunosuppressed.

112. The method of claim 111 , wherein steps (a) - (c) are repeated at least twice a month, at least once a month, at least every 2 months, at least every 3 months, at least every 4 months, at least every 5 months, at least every 6 months, or at least once a year.

113. A biomarker selected from lysine, caffeine, benzoyl formic acid, cytidine, glucuronic acid, glucaric acid, lactose, glucose-6-phosphate, 1 -methylhistamine, histidine, trigonelline, trimethylamine N-oxide (TMAO), hippuric acid, choline, phosphoethanolamine, phosphocholine, acetylglycine, phenyllactic acid, mycophenolic acid glucuronide (MPAG), creatine, and hydroxyphenylacetic acid for use in a method of diagnosing over-immunosuppression or underimmunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy.

114. A kit comprising: a container for collecting a urine sample from a renal transplant recipient, and instructions to halt, alter, or monitor treatment of the renal transplant recipient based on analyzing the levels of one or more metabolites in the urine sample using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed or underimmunosuppressed according to the method of any one of claims 92-112.

115. A computer implemented method for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed, underimmunosuppressed, or stable, the computer performing steps comprising:(a) receiving nuclear magnetic resonance (NMR) and / or mass spectrometry spectra of one or more metabolites from a urine sample obtained from the renal transplant recipient;(b) measuring levels of the one or more metabolites using the NMR spectra and / or mass spectrometry spectra;(c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is over-immunosuppressed, underimmunosuppressed, or stable; and(d) displaying information regarding whether the renal transplant recipient is overimmunosuppressed, under-immunosuppressed, or stable.

116. The computer implemented method of claim 115, further comprising storing the information regarding whether the renal transplant recipient is over-immunosuppressed, underimmunosuppressed, or stable in a database.

117. The computer implemented method of claim 115 or 1 16, wherein the NMR spectra are multi-dimensional NMR spectra.

118. The computer implemented method of claim 1 17, wherein the NMR spectra comprise one-dimensional (1 D)1H NMR spectra, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectra, or a combination thereof.

119. The computer implemented method of claim 117 or 118, wherein the NMR spectra are obtained using non-uniformed sampling (NUS).

120. The computer implemented method of any one of claims 115-119, wherein one or more differential NMR metabolite features are used to classify the renal transplant recipient as overimmunosuppressed, under-immunosuppressed, or stable using the machine learning model, wherein the one or more differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable are selected from X109 having a chemical shift at 3.011H ppm and a chemical shift at 41.9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81.6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm; and wherein the one or more differential NMR metabolite features used to classify the renal transplant recipient as under-immunosuppressed or stable are selected from X65 having a chemical shift at 4.06 ’H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X1223 having a chemical shift at 7.941H ppm and a chemical shift at 140.3013C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.

121. The computer implemented method of claim 120, wherein the differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm, X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, and X1103 having a chemical shift at 3.961H ppm and a chemical shift at 72.8513C ppm.

122. The computer implemented method of claim 120, wherein the differential NMR metabolite features used to classify the renal transplant recipient as over-immunosuppressed or stable comprise or consist of X109 having a chemical shift at 3.011H ppm and a chemical shift at 41 .9713C ppm, X32 having a chemical shift at 4.691H ppm and a chemical shift at 81 .6913C ppm,X40 having a chemical shift at 3.971H ppm and a chemical shift at 76.0513C ppm, X44 having a chemical shift at 4.281H ppm and a chemical shift at 76.2113C ppm, and X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm.

123. The computer implemented method of any one of claims 115-122, wherein the differential NMR metabolite features used to classify the renal transplant recipient as underimmunosuppressed or stable comprise or consist of X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51 .0413C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.

124. The computer implemented method of any one of claims 115-123, further comprising using one or more normalization NMR metabolite features in combination with the differential NMR metabolite features to classify the renal transplant recipient as over-immunosuppressed or stable using the machine learning model.

125. The computer implemented method of claim 124, wherein the one or more normalization NMR metabolite features are selected from X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having a chemical shift at 1 .561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.241H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.

126. The computer implemented method of claim 125, wherein the normalization NMR metabolite features comprise or consist of X205 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X108 having a chemical shift at 2.551H ppm and a chemical shift at 43.1613C ppm, X48 having a chemical shift at 4.081H ppm and a chemical shift at 74.9113C ppm, X11 having a chemical shift at 7.821H ppm and a chemical shift at 129.8913C ppm, X280 having achemical shift at 1.561H ppm and a chemical shift at 28.3113C ppm, X120 having a chemical shift at 1.911H ppm and a chemical shift at 32.7213C ppm, X59 having a chemical shift at 5.24 ’H ppm and a chemical shift at 72.4813C ppm, X63 having a chemical shift at 3.831H ppm and a chemical shift at 71 .9313C ppm, X83 having a chemical shift at 3.281H ppm and a chemical shift at 62.2113C ppm, and X92 having a chemical shift at 3.781H ppm and a chemical shift at 57.0813C ppm.127 The computer implemented method of claim 124, wherein the normalization NMR metabolite features comprise hippuric acid, lysine, mycophenolic acid glucuronide (MPAG), mannitol, trimethylamine N-oxide, and glutamine NMR features.

128. A system for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed or under-immunosuppressed using the computer implemented method of any one of claims 115-127, the system comprising:(a) a storage component for storing data, wherein the storage component has instructions for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed or under-immunosuppressed based on analysis of the NMR spectra of the one or more metabolites stored therein;(b) a computer processor programmed to analyze the NMR spectra using one or more algorithms, wherein the computer processor is coupled to the storage component and configured to execute the instructions stored in the storage component in order to receive the inputted NMR spectra, and analyze the NMR spectra according to the computer implemented method of any one of claims 95-107; and(c) a display component for displaying information regarding whether the renal transplant recipient is over-immunosuppressed or under-immunosuppressed.

129. A non-transitory computer-readable medium comprising program instructions that, when executed by a processor in a computer, causes the processor to perform the computer implemented method of any one of claims 115-127.

130. A kit comprising the non-transitory computer-readable medium of claim 129 and instructions for evaluating whether a renal transplant recipient undergoing treatment with immunosuppressive therapy is over-immunosuppressed or under-immunosuppressed.

131. A method of detecting under-immunosuppression in a renal transplant recipient undergoing treatment with immunosuppressive therapy, the method comprising:(a) obtaining a urine sample from the renal transplant recipient;(b) measuring levels of one or more metabolites in the urine; and(c) analyzing the levels of the one or more metabolites using a machine learning model to determine whether the renal transplant recipient is under-immunosuppressed.

132. The method of claim 131 , wherein the one or more metabolites are selected from glucuronic acid, glucaric acid, lactose, glucose-6-phosphate, 1 -methylhistamine, histidine, trigonelline, trimethylamine N-oxide (TMAO), hippuric acid, choline, phosphoethanolamine, phosphocholine, acetylglycine, phenyllactic acid, mycophenolic acid glucuronide (MPAG), creatine, and hydroxyphenylacetic acid.

133. The method of claim 131 or 132, further comprising altering the immunosuppressive therapy administered to the renal transplant recipient if the renal transplant recipient is determined to be under-immunosuppressed, wherein immunosuppression is increased if the renal transplant recipient is determined to be under-immunosuppressed.

134. The method of any one of claims 131-133, further comprising performing diagnostic screening to determine whether the renal transplant recipient has graft dysfunction, subclinical graft rejection, or graft rejection if the renal transplant recipient is determined to be underimmunosuppressed; and treating the renal transplant recipient for the graft dysfunction, subclinical graft rejection, or graft rejection.

135. The method of any one of claims 131 -134, wherein the machine learning model uses an efficient neural network (ENET), an orthogonal projections to latent structures-discriminant analysis (OPLS-DA) or a random forest (RF) machine learning algorithm.

136. The method of any one of claims 131-135, wherein said measuring comprises performing nuclear magnetic resonance (NMR) spectroscopy or mass spectrometry.

137. The method of claim 136, wherein the NMR spectroscopy is multi-dimensional NMR spectroscopy.

138. The method of claim 137, wherein the NMR spectroscopy comprises one-dimensional (1 D) ’H NMR spectroscopy, 2-dimensional (2D)13C-1H heteronuclear single quantum coherence (HSQC) NMR spectroscopy, or a combination thereof.

139. The method of claim 137 or 138, wherein the NMR is performed with non-uniformed sampling (NUS).

140. The method of any one of claims 136-139, wherein one or more differential NMR metabolite features are used to classify the renal transplant recipient as under-immunosuppressed or stable using the machine learning model, wherein the one or more differential NMR metabolite features are selected from X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X1223 having a chemical shift at 7.941H ppm and a chemical shift at 140.3013C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, X93 having a chemical shift at 4.081H ppm and a chemical shift at 74.8313C ppm, X129 having a chemical shift at 3.281H ppm and a chemical shift at 62.2313C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.

141. The method of claim 140, wherein the differential NMR metabolite features used to classify the renal transplant recipient as under-immunosuppressed or stable comprise or consist of X65 having a chemical shift at 4.061H ppm and a chemical shift at 85.9413C ppm, X56 having a chemical shift at 4.671H ppm and a chemical shift at 98.5713C ppm, X360 having a chemical shift at 2.171H ppm and a chemical shift at 24.5113C ppm, X333 having a chemical shift at 4.441H ppm and a chemical shift at 51.0413C ppm, and X161 having a chemical shift at 3.971H ppm and a chemical shift at 46.5513C ppm.

Images