Systems and methods for evaluation of circadian rhythm and sleep homeostasis

Abstract

Systems and methods for assessment of circadian rhythm and sleep homeostasis are described. Presence of proteinaceous biomarkers within an individual's biological sample can be utilized in a computational classifier to indicate rhythmic phase and / or sleep homeostasis. Further clinical analysis and / or treatments can be performed based on determined rhythmic phase or sleep homeostasis.

Description

SYSTEMS AND METHODS FOR EVALUATION OF CIRCADIAN RHYTHM ANDSLEEP HOMEOSTASISCROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims benefit of U.S. Provisional Patent Application No. 63 / 600,597, filed November 17, 2023, entitled “Systems and Methods for Evaluation of Circadian Rhythm and Diurnal Rhythm,” the disclosure of which is hereby incorporated by reference in its entirety for all purposes.TECHNOLOGICAL FIELD

[0002] The disclosure is generally directed to processes to evaluate circadian rhythm, sleep homeostasis, and sleep disorders of individuals, which may be utilized in clinical assessments, monitoring, and / or interventions.BACKGROUND

[0003] Circadian rhythms and sleep homeostasis are powerful regulators of physiology. Although often conflated, circadian rhythm and sleep homeostasis are two different phenomena affecting sleep. Circadian rhythm is the process of daily oscillations, timing cyclic physiological events such as the release of melatonin to induce sleep. Sleep homeostasis, on the other hand, is the process regulating the amount sleep and restfulness and will induce tiredness and sleep when deprived of sleep.

[0004] In mammals, circadian rhythms are driven by a central pacemaker located in the suprachiasmatic nucleus (SCN), which receives direct input from the visual system, allowing synchronization to the external 24-hour day. The central clock, in turn, is connected through neuronal and endocrine networks to other tissues which have their own endogenous circadian clocks, allowing each system to coordinate temporal programs optimal to the organism. Although much is known regarding how each cell generates its own circadian rhythm through a process involving transcriptional / translational feedback loops, the synchronization between peripheral and central clocks is poorly understood, and there is evidence these systems can become dissociated. As an example, in rodents and humans, restricting food availability can shiftthe phase of metabolic processes to times when food is available, with deleterious metabolic consequences (J. L. Barclay, et al., PLoS One. 2012; 7 (5): e37150; and D. B. Boivin, et al., Journal of biological rhythms. 2022; 37 (1 ): 3-28; the disclosures of which are incorporated herein by reference).

[0005] Sleep homeostasis also regulates physiology in the entire body. When the amount of sleep deprivation (also referred to as sleep debt) increases beyond a threshold, inducing a variety of physiological and neurobehavioral changes resulting in fatigue (e.g., reduced vigilance and psychomotor speed) and desire for sleep. Sleep deprivation can be acute (e.g., no sleep the night before) or chronic (e.g., sleeping only a few hours each night over a week), but the neurobehavioral effects of chronic sleep deprivation are less severe than those observed after acute total sleep deprivation.

[0006] The interaction of sleep homeostasis and circadian regulation is best summarized by the “two process mode!” (Goel N, Basner M, Rao H, Dinges DF, Prog Mol Biol Transl Sci. 2013;119:155-90; and Borbely, J Sleep Res. 2022 Aug;31 (4):e13598, the disclosures of which are hereby incorporated by reference). Unfortunately, assessment of these physiological process is quite difficult. Currently, there are no existing reliable biomarkers of sleep homeostasis or sleep deprivation. And measuring circadian phase is cumbersome. The standard and most common way to assess circadian phase is a test called the Dim Light Melatonin Onset (DLMO), which involves having the patient in a controlled environment for an evening with tightly controlled light exposure mimicking the daily sunset and measuring melatonin in 10-20 successive blood or saliva samples (Kennaway DJ. J Pineal Res. 2020 Aug;69(1 ):e12657). For this reason, there is interest in understanding and describing how these processes are organized and finding better biomarkers of these physiological processes.SUMMARY

[0007] Several embodiments of the disclosure are directed towards development and use of one or more computational models to predict circadian phase and / or sleep homeostasis status. Generally, measurements of proteinaceous species are received, which can be obtained from a collection of a biological sample of an individual. Based on the timing of the sample collection and the results a one or more computational modelstrained to predict circadian phase and / or sleep homeostasis status, the individual can be assessed for offset circadian rhythm, sleep deprivation, and disorders related to circadian rhythm and sleep deprivation. In some implementations, the assessment of proteinaceous species within the computational model for predicting circadian rhythm is used a surrogate for the Dim Light Melatonin Onset (DLMO) test. The assessment of proteinaceous species within the computational model for predicting sleep homeostasis is the first practical assessment of sleep deprivation. Use of the one or more computational models provides for various applications, including diagnosis and treatments of sleep disorders. It can further differentiate subtypes of sleep disorders to better inform treatment options. And furthermore, a number of medications have impaired efficacy based on timing of administration and the recipient’s circadian phase. One such class of medications with efficacy affected by circadian phase are immune checkpoint inhibitors for treatment of cancer. Accordingly, when an individual is to be treated with an immune checkpoint inhibitor, the recipient is assessed to determine circadian phase and treatment is aligned with the individual’s circadian clock and / or is further treated with a medication for regulating circadian rhythm.

[0008] In some aspects, the techniques described herein relate to a method of predicting circadian rhythmic phase of an individual, including: receiving, using a computational processing system, proteinaceous species measurements from a biological sample collected from an individual; and classifying, using the computational processing system, the biological sample as being associated with a rhythmic phase by entering the proteinaceous species measurements into a trained rhythmic phase classification model.

[0009] In some aspects, the techniques described herein relate to a method further including: determining a timing of a circadian rhythmic event of the individual based on the time the biological sample was collected and the rhythmic phase of the biological sample as classified by the trained rhythmic phase classification model.

[0010] In some aspects, the techniques described herein relate to a method, wherein the circadian rhythmic event is determined to be delayed or advanced, the method further including: administering one or more modulators of circadian rhythm phase.

[0011] In some aspects, the techniques described herein relate to a method, wherein the one or more modulators modulator or circadian phase includes one or more melatonin receptor agonists or light therapy.

[0012] In some aspects, the techniques described herein relate to a method, wherein the circadian rhythmic event is dim light melatonin onset.

[0013] In some aspects, the techniques described herein relate to a method further including: correcting for individualism using the computational processing system, wherein correcting for individualism includes: determining a proteinaceous species signature of the individual; determining, using the proteinaceous species signature, circadian-rhythmic-related species measurements deviated from a population norm; and removing the deviated circadian-rhythmic-related species prior to entering the proteinaceous species measurements into the trained rhythmic phase classification model.

[0014] In some aspects, the techniques described herein relate to a method further including: correcting for individualism using the computational processing system, wherein correcting for individualism includes: determining a proteinaceous species signature of the individual; determining an individual's personalized center of the proteinaceous species signature; and adjusting the classification result using the individual's personalized center based on a difference between a center of a population norm and the individual's personalized center.

[0015] In some aspects, the techniques described herein relate to a method further including: measuring proteinaceous species within a proteinaceous species sample, wherein the proteinaceous species sample is or is derived from the biological sample.

[0016] In some aspects, the techniques described herein relate to a method, wherein measuring proteinaceous species includes one or more of the following: ELISA, antibody arrays, DNA bar-coded antibodies, aptamers, chromatography, mass spectrometry, NMR, or electrophoresis.

[0017] In some aspects, the techniques described herein relate to a method, wherein measuring proteinaceous species includes: contacting the proteinaceous species sample with a panel of aptamers that are configured to bind select proteinaceous species; and sequencing the aptamers that bind a proteinaceous species.

[0018] In some aspects, the techniques described herein relate to a method, wherein measuring proteinaceous species includes: contacting the proteinaceous species sample with a panel of DNA bar-coded antibodies that are configured to bind select proteinaceous species; and sequencing DNA bar codes of the antibodies that bind a proteinaceous species.

[0019] In some aspects, the techniques described herein relate to a method, wherein the aptamers or the DNA bar-coded antibodies are configured to bind two or more of the following proteinaceous species: angiopoietin-related protein 1 , cathepsin F, proopiomelanocortin, prolactin, metalloproteinase inhibitor 4, hyaluronidase-1 , thyroid stimulating hormone, kallikrein-7, secretogranin-1 , cysteine-rich secretory protein LCCL domain-containing 2, hemopexin, zymogen granule protein 16 homolog B, pancreatic hormone, follistatin-related protein 3, neutrophil cytosol factor 1 , heparan-sulfate 6-0- sulfotransferase 2, cGMP-dependent 3',5'-cyclic phosphodiesterase, HLA class I alpha chain G, lactotransferrin, proprotein convertase subtilisin / kexin type 9, secreted and transmembrane protein 1 , desmocollin-2, succinate dehydrogenase assembly factor 2- mitochondrial, and lysosomal protective protein.

[0020] In some aspects, the techniques described herein relate to a method wherein the trained rhythmic phase classification model includes the use of one or more of: linear regression, polynomial regression, Cox proportional hazards regression, multiple linear regression, ridge regression, logistic regression, Lasso regression, stepwise regression, principal component analysis, Bayesian inference, elastic net, and random forest regression.

[0021] In some aspects, the techniques described herein relate to a method further including: training a classifier to yield the trained rhythmic phase classification model, including: collecting a plurality of biological samples for each individual of a cohort of individuals, wherein the plurality of biological samples span across multiple circadian phases; determining a circadian rhythmic event for each individual of the cohort; assigning each biological sample a time stamp relative to the circadian rhythmic event; measuring proteinaceous species within each biological sample to yield proteinaceous species measurements; and entering the proteinaceous species measurements and associated circadian rhythmic event time stamp into a classifier model to train the modelto learn the relation between the proteinaceous species measurements and the circadian rhythmic event time stamp.

[0022] In some aspects, the techniques described herein relate to a method, wherein the circadian rhythmic event is dim light melatonin onset.

[0023] In some aspects, the techniques described herein relate to a method, wherein the individual has a medical condition selected from one of: one of blindness, narcolepsy, chronic insomnia, hypersomnia, advanced sleep-wake phase disorder, delayed sleepwake phase disorder, irregular sleep-wake rhythm disorder, jet lag disorder, non-24-hour sleep-wake rhythm disorder, or shift work disorder.

[0024] In some aspects, the techniques described herein relate to a method of improving administration of a treatment affected by circadian rhythm phase, including: receiving, using a computational processing system, measurements of proteinaceous species from a biological sample collected from an individual, wherein the individual has diagnosis for a medical condition to be with treated with a medication that is affected by circadian rhythm; classifying, using the computational processing system, the biological sample as being associated with a rhythmic phase by entering the proteinaceous species measurements into a trained rhythmic phase classification model; determining a timing of a circadian rhythmic event of the individual based on the time the biological sample was collected and the rhythmic phase of the biological sample as classified by the trained rhythmic phase classification model; and administering the medication to the individual to treat the medical condition at an efficacious time based on the individual's determined timing of the circadian rhythmic event.

[0025] In some aspects, the techniques described herein relate to a method further including: administering one or more modulators of circadian rhythm phase to modulate the timing of the circadian rhythmic event.

[0026] In some aspects, the techniques described herein relate to a method, wherein the one or more modulators modulator or circadian phase includes one or more melatonin receptor agonists or light therapy.

[0027] In some aspects, the techniques described herein relate to a method, wherein the circadian rhythmic event is dim light melatonin onset.

[0028] In some aspects, the techniques described herein relate to a method further including: correcting for individualism using the computational processing system, wherein correcting for individualism includes: determining a proteinaceous species signature of the individual; determining, using the proteinaceous species signature, circadian-rhythmic-related species measurements deviated from a population norm; and removing the deviated circadian-rhythmic-related species prior to entering the proteinaceous species measurements into the trained rhythmic phase classification model.

[0029] In some aspects, the techniques described herein relate to a method further including: correcting for individualism using the computational processing system, wherein correcting for individualism includes: determining a proteinaceous species signature of the individual; determining an individual's personalized center of the proteinaceous species signature; and adjusting the classification result using the individual's personalized center based on a difference between a center of a population norm and the individual's personalized center.

[0030] In some aspects, the techniques described herein relate to a method further including: measuring proteinaceous species within a proteinaceous species sample, wherein the proteinaceous species sample is or is derived from the biological sample.

[0031] In some aspects, the techniques described herein relate to a method, wherein measuring proteinaceous species includes one or more of the following: ELISA, antibody arrays, DNA bar-coded antibodies, aptamers, chromatography, mass spectrometry, NMR, or electrophoresis.

[0032] In some aspects, the techniques described herein relate to a method, wherein measuring proteinaceous species includes: contacting the proteinaceous species sample with a panel of aptamers that are configured to bind select proteinaceous species; and sequencing the aptamers that bind a proteinaceous species.

[0033] In some aspects, the techniques described herein relate to a method, wherein measuring proteinaceous species includes: contacting the proteinaceous species sample with a panel of DNA bar-coded antibodies that are configured to bind select proteinaceous species; and sequencing DNA bar codes of the antibodies that bind a proteinaceous species.

[0034] In some aspects, the techniques described herein relate to a method, wherein the aptamers or the DNA bar-coded antibodies are configured to bind two or more of the following proteinaceous species: angiopoietin-related protein 1 , cathepsin F, proopiomelanocortin, prolactin, metalloproteinase inhibitor 4, hyaluronidase-1 , thyroid stimulating hormone, kallikrein-7, secretogranin-1 , cysteine-rich secretory protein LCCL domain-containing 2, hemopexin, zymogen granule protein 16 homolog B, pancreatic hormone, follistatin-related protein 3, neutrophil cytosol factor 1 , heparan-sulfate 6-0- sulfotransferase 2, cGMP-dependent 3',5’-cyclic phosphodiesterase, HLA class I alpha chain G, lactotransferrin, proprotein convertase subtilisin / kexin type 9, secreted and transmembrane protein 1 , desmocollin-2, succinate dehydrogenase assembly factor 2- mitochondrial, and lysosomal protective protein.

[0035] In some aspects, the techniques described herein relate to a method wherein the trained rhythmic phase classification model includes the use of one or more of: linear regression, polynomial regression, Cox proportional hazards regression, multiple linear regression, ridge regression, logistic regression, Lasso regression, stepwise regression, principal component analysis, Bayesian inference, elastic net, and random forest regression.

[0036] In some aspects, the techniques described herein relate to a method, wherein the medical condition includes cancer and the medication includes an immune checkpoint inhibitor.

[0037] In some aspects, the techniques described herein relate to a method, wherein the medical condition includes cancer and the medication includes a chemotherapeutic.

[0038] In some aspects, the techniques described herein relate to a method, wherein the medical condition includes high cholesterol and the medication includes a statin.

[0039] In some aspects, the techniques described herein relate to a method, wherein the medical condition includes hypertension and the medication includes an alphablocker.

[0040] In some aspects, the techniques described herein relate to a method, wherein the medical condition includes hypertension and the medication includes a beta-blocker.

[0041] In some aspects, the techniques described herein relate to a method, wherein the medical condition includes hypertension and the medication includes an angiotensin II receptor blocker.

[0042] In some aspects, the techniques described herein relate to a method, wherein the medical condition includes allergic dermatitis or rhinitis and the medication includes an antihistamine.

[0043] In some aspects, the techniques described herein relate to a method, wherein the medical condition includes arthritis and the medication includes an NSAID.

[0044] In some aspects, the techniques described herein relate to a method, wherein the medical condition includes arthritis and the medication includes an antimetabolite.

[0045] In some aspects, the techniques described herein relate to a method, wherein the medical condition includes asthma and the medication includes a phosphodiesterase inhibitor.

[0046] In some aspects, the techniques described herein relate to a method, wherein the medical condition includes clotting and the medication includes a p2y12 inhibitor.

[0047] In some aspects, the techniques described herein relate to a method, wherein the medical condition includes anticoagulation and the medication includes a Factor Xa inhibitor.

[0048] In some aspects, the techniques described herein relate to a method, wherein the medical condition includes gastroesophageal reflux disease and the medication includes a proton pump inhibitor.

[0049] In some aspects, the techniques described herein relate to a method of identifying a subtype of delayed sleep-wake phase disorder, including: receiving, using a computational processing system, measurements of proteinaceous species from a biological sample collected from an individual, wherein the individual has diagnosis for delayed sleep-wake phase disorder; classifying, using the computational processing system, the biological sample as being associated with a rhythmic phase by entering the proteinaceous species measurements into a trained rhythmic phase classification model; determining a timing of a dim light melatonin onset of the individual based on the time the biological sample was collected and the rhythmic phase of the biological sample as classified by the trained rhythmic phase classification model; and determining that theindividual is a circadian delayed subtype or a non-circadian subtype based on the timing of a dim light melatonin onset.

[0050] In some aspects, the techniques described herein relate to a method further including: determining that the individual is a circadian delayed subtype; and administering one or more modulators of circadian rhythm phase to modulate the timing of the dim light melatonin onset.

[0051] In some aspects, the techniques described herein relate to a method, wherein the one or more modulators modulator or circadian phase includes one or more melatonin receptor agonists or light therapy.

[0052] In some aspects, the techniques described herein relate to a method further including: determining that the individual is a non-circadian subtype, wherein the individual is not administered one or more modulators of circadian rhythm phase to modulate the timing of the dim light melatonin onset.

[0053] In some aspects, the techniques described herein relate to a method further including: administering a hypnotic agent to the individual.

[0054] In some aspects, the techniques described herein relate to a method, wherein the hypnotic agent includes one or more of: a benzodiazepine, a non-benzodiazepine sedative-hypnotic, an orexin receptor antagonist, a sedative antidepressant, or an antihistamine.

[0055] In some aspects, the techniques described herein relate to a method further including: correcting for individualism using the computational processing system, wherein correcting for individualism includes: determining a proteinaceous species signature of the individual; determining, using the proteinaceous species signature, circadian-rhythmic-related species measurements deviated from a population norm; and removing the deviated circadian-rhythmic-related species prior to entering the proteinaceous species measurements into the trained rhythmic phase classification model.

[0056] In some aspects, the techniques described herein relate to a method further including: correcting for individualism using the computational processing system, wherein correcting for individualism includes: determining a proteinaceous species signature of the individual; determining an individual's personalized center of theproteinaceous species signature; and adjusting the ciassification result using the individual's personalized center based on a difference between a center of a population norm and the individual's personalized center.

[0057] In some aspects, the techniques described herein relate to a method further including: measuring proteinaceous species within a proteinaceous species sample, wherein the proteinaceous species sample is or is derived from the biological sample.

[0058] In some aspects, the techniques described herein relate to a method, wherein measuring proteinaceous species includes one or more of the following: ELISA, antibody arrays, DNA bar-coded antibodies, aptamers, chromatography, mass spectrometry. NMR, or electrophoresis.

[0059] In some aspects, the techniques described herein relate to a method, wherein measuring proteinaceous species includes: contacting the proteinaceous species sample with a panel of aptamers that are configured to bind select proteinaceous species; and sequencing the aptamers that bind a proteinaceous species.

[0060] In some aspects, the techniques described herein relate to a method, wherein measuring proteinaceous species includes: contacting the proteinaceous species sample with a panel of DNA bar-coded antibodies that are configured to bind select proteinaceous species; and sequencing DNA bar codes of the antibodies that bind a proteinaceous species.

[0061] In some aspects, the techniques described herein relate to a method, wherein the aptamers or the DNA bar-coded antibodies are configured to bind two or more of the following proteinaceous species: angiopoietin-related protein 1 , cathepsin F, proopiomelanocortin, prolactin, metalloproteinase inhibitor 4, hyaluronidase-1 , thyroid stimulating hormone, kallikrein-7, secretogranin-1 , cysteine-rich secretory protein LCCL domain-containing 2, hemopexin, zymogen granule protein 16 homolog B, pancreatic hormone, follistatin-related protein 3, neutrophil cytosol factor 1 , heparan-sulfate 6-O- sulfotransferase 2, cGMP-dependent 3',5'-cyclic phosphodiesterase, HLA class I alpha chain G, lactotransferrin, proprotein convertase subtilisin / kexin type 9, secreted and transmembrane protein 1 , desmocollin-2, succinate dehydrogenase assembly factor 2- mitochondrial, and lysosomal protective protein.

[0062] In some aspects, the techniques described herein relate to a method wherein the trained rhythmic phase classification model includes the use of one or more of: linear regression, polynomial regression, Cox proportional hazards regression, multiple linear regression, ridge regression, logistic regression, Lasso regression, stepwise regression, principal component analysis, Bayesian inference, elastic net, and random forest regression.BRIEF DESCRIPTION OF THE DRAWINGS

[0063] The description and claims will be more fully understood with reference to the following figures and data graphs, which are presented as exemplary embodiments of the invention and should not be construed as a complete recitation of the scope of the invention.

[0064] Figure 1 provides an example of a method to classify a sample based on proteinaceous species.

[0065] Figure 2 provides a schematic of an example of a computational system to perform various computational methods.]

[0066] Figures 3A, 3B and 3C provide an overview of study protocols and examples of melatonin profiles from each study condition. Fig. 3A: schematic of study protocols from Study 1 and Study 2. Relative clock times of study events, lighting levels (white-room light, grey-dim light, black=lights out / dark), sleep and wake times, study sampling segments (in Study 1 blue=sampling from the 24-hour sleep-wake day; orange=sampling from the first 24 hours of CR; purple=sampling from the final 15 hours of CR and recovery night: in Study 2 green^first 24 hours of sampling from CR), and approximate timing of proteomic samples and melatonin samples from each segment. Lower panel: Data from one exemplar participant in Study 1 is shown in Fig. 3B and data from one participant in Study 2 is shown in Fig. 3C; background colors correspond to study sampling segments shown in Fig. 3A. Black dots indicate samples for which both melatonin and proteomics analysis was carried out, grey dots indicate sample times when only melatonin was measured. The 3-harmonic fit to the melatonin data is shown in the grey curve, and the vertical dashed line indicates the time at which the rising melatonin reached 25% of thefundamental harmonic level, denoted as Dim Light Melatonin Onset (DLMO). The arrows with clock time indicate the clock time of DLMO,

[0067] Figure 4 provides data on tuning hyperparameters to balance specificity and sensibility. Rhythmic proteins were identified based on two criteria: the ability to satisfy the false discovery rate (FDR) rhythm threshold (displayed in the left panel) and the amplitude requirement (shown in the middle panel) in at least one condition. The amplitude was determined using the 2-norm of its components. Each hyperparameter was investigated separately while keeping other parameters at default values. Additionally, proteins that showed a rhythmic pattern and did not have a significant difference in the pattern across the different conditions (based on the FDR comparison threshold), were categorized as circadian (shown in the right panel). 1 -harmonic (dot curve) and 2-harmonic (dot-dash curve) results were merged by taking the union (full curve). The default parameter values used in our study were rhy_fdr = 0.01, amp_cutoff ~ 0.1, and com_fdr ~ 0.05.

[0068] Figures 5A and 5B provide data of expression patterns of selected proteins. Fig. 5A shows examples of four rhythmic proteins detected using the 2-harmonic approach, while Fig. 5B showcases the eight most robust circadian proteins. The individual curves are derived through a series of steps, including log-normalization, meanfiltering, and soft-dtw barycenter computation. X-axis is the time relative to Dim Light Melatonin Onset (DLMO). protein pattern from the 24-hour sleep-wake day in Study 1 ; protein pattern from the first 24 hours of CR in Study 1 ; protein pattern from the final 15 hours of CR and recovery night in Study 1 ; protein pattern from the first 24 hours of CR in Study 2. Transparent curves represent variations computed using the same process but deviating from the main barycenter by a single subject. Note that not all proteins were assessed in Study 2, which was tested with a different version of the Somalogic panel that did not include them. Additionally, SomaLogic sometimes uses two types of aptamers for the same protein, which is indicated here by a sequence id at the end of the protein’s name for Anterior gradient protein 3 in Fig. 5A.

[0069] Figure 6 provides data on pathway analysis for endogenous circadian proteins with respect to peak times. Pathway analysis was carried out on circadian proteins peaking within ±1 hour with GO__Biological_Process__2021 , Reactome__2022,KEGG_2021_Human database using the Gene Set Enrichment Analysis (GSEA) computational method. The combined score was normalized along the time axis to visualize when the pathway is the most active. The solid black line illustrates the number of endogenous circadian proteins peaking at each time (right axis) used to conduct the pathway analysis. The vertical dashed lines delineate the two times of day when most circadian protein peaks occurred. Relative clock hour is presented along the top x-axis, while time relative to Dim Light Melatonin Onset (DLMO) is presented along the bottom x-axis.

[0070] Figure 7 provides a schematic of Somascan. A) SOMAmers attached to beads bind to proteins in the sample mix. B) Proteins that are bound to their specific SOMAmer reagents are biotinylated. C) The SOMAmer-protein complexes are released by a photocleavable linker. D) The biotinylated proteins are bound to a second streptavidin bead. E) Where the bound SOMAmer reagents are removed from their protein targets. F) Released SOMAmer reagents are collected, denatured, and quantified using a custom microarray.

[0071] Figures 8A and 8B provide data on performance metrics of the Somascan circadian time predictor. Fig. 8A: This model was used to predict the circadian time of samples in 2 validations. While datasets called Harvard validation and Stanford validation were assayed together, the validation cohort called Depner et al 2018 was a published dataset assayed independently. x~axis is true clock hr of the sample, y-axis is the predicted clock hr, MdAE is the median absolute error computed as the median difference between predicted hr and true hr. Fig. 8B: Cumulative frequency distributions of the absolute prediction errors of the model in the two validation cohorts. The median and the 80% quantile of the errors are marked respectively.

[0072] Figure 9 provides a heat map of a circadian timestamp constructed by fitting harmonic components from 5500 plasma proteins assayed by the Somascan platform (379 proteins significant at FDR p value = 0.05). This plot depicts the mean protein expression levels of 9 individuals who were sampled every 2 h over a 35 h interval under the controlled behavioral and environmental conditions of a Constant Routine. In the inset marked in white are the classic circadian proteins ACTH and Pro-opiomelanocortin that peak in the early hours of the day.

[0073] Figure 10 provides data on performance metrics of the Somascan 5500 protein panel trained on 100 timepoints and validated in an untouched validation dataset (n=53). 119 proteins were deemed as important coefficients in the lasso model, x-axis is true DLMO hr of the sample, y-axis is the predicted DLMO hr, MdAE is the median absolute error computed as the median difference between predicted hr and true hr, r is circular pearson’s correlation coefficient of true and predicted time and its corresponding Pearson’s circular correlation p value. Cumulative frequency distributions of the absolute prediction errors of the model in the validation cohort (n=53 timepoints). The median and the 80% quantile of the errors are marked respectively. Note that prediction is identical by sex.

[0074] Figure 11 provides a table of the most frequent proteins (at least 70% of 100 instances) selected in training the lasso based circadian time predictor PlasmaTime on the Somascan 5500 protein panel.

[0075] Figure 12 provides data comparing DLMO performed at home vs in the lab. Timing of dim light melatonin onset (DLMO) from samples collected by participants in their home vs. the following evening in a laboratory setting. Open symbols: young participants with a complaint of difficulty falling asleep or difficulty waking; filled symbols: older participants with a complaint of difficulty falling asleep too early or waking too early; dashed line: line of unity. The two measures were highly correlated (r=0.85, n=18, p<0.0001 ).

[0076] Figure 13 provides data of DLMO captured from a patient with delayed sleepwake phase disorder with samples collected at home.DETAILED DESCRIPTION

[0077] Turning now to the drawings and data, systems and methods for assessment of circadian rhythm, sleep homeostasis, and sleep disorders of an individual and applications thereof are described. In some embodiments, proteinaceous analyte measurements are utilized within a trained computational model to predict rhythmic phase and / or sleep homeostasis status. The proteinaceous analytes that are measured are derived from a biological sample (e.g., blood or plasma). In some embodiments, the biological sample is collected from an individual to be assessed for rhythmic phase and / orsleep homeostasis status. The time the sampie is collected can be utilized to determine whether the rhythmic phase is in line with the expected rhythmic phase at the time of sampling. If the rhythmic phase is offset and / or the sleep homeostasis status yields an indication of sleep deprivation, the resulting analysis can help infer whether the individual from which the sample was collected is experiencing abnormalities in circadian rhythm, abnormalities in diurnal rhythm, experiencing sleep deprivation, and / or has a sleep disorder (or a subtype of sleep disorder). In some embodiments, based on the proteinaceous analyte analysis results indicating a rhythmic abnormality, sleep deprivation, or presence of a sleep disorder, an intervention is performed, such as further clinical assessment and / or treatment. In some embodiments, the proteinaceous analyte analysis results yield an indication of the type of treatment to be performed (or not to be performed). In some embodiments, the proteinaceous analyte analysis is performed to determine how to administer a medication in which the efficacy of the medication is affected by rhythmic phase.

[0078] Traditionally, timing (or phase) of the human central circadian clock is measured using the Dim Light Melatonin Onset (DLMO). which in essence assesses secretion of melatonin by the pineal gland is controlled via a multi-synaptic pathway from the SCN through the superior cervical ganglion (J. Arendt, J Biol Rhythms. 2005; 20 (4): 291-303; and S. Benloucif, et al., J Clin Sleep Med. 2008; 4 (1): 66-69; the disclosures of which are incorporated herein by reference). Melatonin is at very low or non-detectable levels in blood and saliva during the daytime and begins secretion in the evening, reaching peak levels in the middle of the night. While the rhythm of melatonin has served a useful purpose for assessing the phase of the central clock for research purposes, it does have several drawbacks. First, melatonin is acutely suppressed by light, and assessment of melatonin in saliva or blood requires the participant to remain in dim light throughout the sampling segment. In addition, measurement from saliva requires an awake participant, while measurement from blood requires either multiple needle sticks (again waking the participant) or a special blood sampling system that allows sampling without waking the participant. Furthermore, these assessments are generally performed in the evening (e.g., when the sun sets), take several hours, and require the patient to be in facility with strictly controlled light conditions. These issues, together with the durationof melatonin sampling required to obtain a phase estimate have severely limited the assessment of melatonin timing in patients. Furthermore, while melatonin provides useful information about central circadian phase, it does not provide any indication of the phase of peripheral oscillators, nor whether peripheral rhythms are synchronized to the central clock or to each other.

[0079] For these reasons, researchers have sought alternatives for assessing endogenous circadian phase in humans. In animals, tissues can be collected and analyzed for metabolites, gene expression, or protein content to assess the phase of both the central and peripheral oscillators. Such studies have led to a wealth of information on how organ clocks are organized, with the limitations that most animal models are nocturnal, inbred, and live in artificial lighting and environmental conditions. Animal models have shown that by measuring multiple parameters at a given time point and considering a stable relationship between phases of various processes, it is possible to derive the central circadian phase with a single time point, a concept called a "time stamp” (H. R. Ueda, et al., Proc Natl Acad Sci U S A. 2004; 101 (31 ): 11227-11232, the disclosure of which is incorporated herein by reference). Subsequently, attempts have been made at establishing time stamp measures that could replace melatonin phase in humans, but in most cases to achieve accuracy within 1-2 hours of DLMO phase at least two biological samples collected hours apart have been needed (see E. E. Laing, et al., Elife. 2017; 6, the disclosure of which is incorporated herein by reference). One exception is BodyTime, a measure of circadian phase from gene expression in isolated monocytes derived from a single blood sample, which has shown good performance in studies against DLMO (N. Wittenbrink, et al., J Clin Invest. 2018; 128 (9): 3826-3839, the disclosure of which is incorporated herein by reference). Although powerful, BodyTime measures the circadian clock in a specific immune tissue that can be perturbed when the immune system is stressed, such as when fighting a pathogenic infection.

[0080] In regards to sleep homeostasis, there are no clinical assessments that are practiced. Biomarkers have yet to be revealed and thus there is no biological-based assessment to identify and / or quantify sleep deprivation. And because there are differences in the amount of sleep an individual requires to achieve sleep sufficiency, it can be difficult to assess by mere quantification of actual hours slept. Sleep deprivation,however, is attributed to affecting brain development in adolescents, worsening many chronic diseases, increasing risk of dementia, and development or worsening of mental health conditions such as depression. It thus vital to be able to assess individuals’ personal sleep homeostasis on a biological level to improve patient care.

[0081] Here, the circulating plasma proteome was assessed to describe proteins that are associated with circadian rhythms and / or sleep homeostasis to develop a single sample proteomic measure for assessing rhythmic phase and / or sleep homeostasis status via a trained computational model. Indeed, the results show that a large number of proteins can be measured in a small amount of blood. In some implementations, proteinaceous species are measured using a DNA bar-coded antibody-based technology (e.g., the Olink platform, Olink Holdings, Uppsala, Sweden) or an aptamer-based technology (e.g., the Somalogic platform, Somalogic, Boulder, CO) (for more on these platforms, see G. S. Omenn, et al., J Proteome Res. 2023; 22 (4): 1024-1042, the disclosure of which is incorporated herein by reference). And in some implementations, more traditional technologies, such as (for example) mass spectrometry or antibody arrays, are utilized to measure proteinaceous species. One benefit of assessing proteinaceous species is that proteins are the structural and acting component of the human body and thus more directly affect human physiology than nucleic acid species. Another advantage of measuring proteinaceous species within the plasma is that it will capture secreted proteins and hormones that are used in inter-organ communications, connecting central and peripheral clocks.

[0082] Based on these findings, several embodiments of the disclosure are directed to systems and methods for assessing rhythmic phase and / or sleep homeostasis / sleep deprivation using a biological sample of an individual and a computational classifier. In many embodiments, the systems and methods for assessing rhythmic phase can be utilized as a surrogate for DLMO. Further, systems and methods for assessing sleep homeostasis / sleep debt provide a means for determining whether an individual is getting enough sleep. These systems and methods allow for easier, better, and a more complete assessment of an individual’s sleep, which can further assist in determining better treatments and behavioral changes to improve the individual’s wellbeing.Classificafion based on proteinaceous biomarkers

[0083] Several embodiments of the disclosure are directed to rhythmic phase and / or sleep homeostasis status classification of a biological sample of an individual. As is described in the various examples of the disclosure, a number of proteinaceous species have been found to be associated with rhythmic phase and / or sleep homeostasis, in which measurements of the proteinaceous species can be utilized within a classifier to predict rhythmic phase and / or sleep homeostasis. In some embodiments, proteinaceous species measurements derive from an individual’s biological sample of are utilized to determine the rhythmic phase of the sample, which can be used to determine the individual’s DLMO. In a similar manner, in some embodiments, proteinaceous species measurements derived from an individual’s biological sample is utilized to determine the sleep homeostasis status of the individual, which can be useful to determine if the individual is sleep deprived.

[0084] An example of a method for classifying rhythmic phase or sleep homeostasis / sleep debt from a sample of proteinaceous species is provided in Fig. 1 . In this example, a biological sample is collected from an individual. Proteinaceous species of the biological sample are measured. Measurements of the proteinaceous species are utilized in a computational classifier to classify the sample to be of a particular rhythmic phase or a particular sleep homeostasis status.

[0085] Method 100 can begin with receiving (101 ) measurements of proteinaceous species, which can be measured from a sample. The sample from which the proteinaceous species are measured can be derived from or is the biological sample that is collected from an individual. Various biological samples can be collected for measurement of proteinaceous species, including (but not limited to) blood, plasma, cerebral spinal fluid, urine, or saliva. It has been found that circulating proteinaceous species are useful in assessment of rhythmic phase or sleep debt. Accordingly, in some particular embodiments, the proteinaceous species that are measured are circulating proteinaceous species, which can be of a blood sample or a plasma sample collected from an individual.

[0086] In some embodiments, the method further comprises collecting the biological sample from the individual. Any appropriate methodology for collecting a biologicalsample can be utilized, such as (for example) a blood draw, a collection of urine, a collection of spit, and a swab of biological fluids. Generally, a biological sample can be collected at any time of day. In several implementations of the method, the time of sample collection is recorded, which may be useful for determining rhythmic phase and / or sleep homeostasis in association with a moment in time. The collected sample can be further processed in order to perform measurement of the proteinaceous species.

[0087] In some implementations, a single biological sample is collected for measuring proteinaceous species, which is sufficient for performing downstream classification of the sample for rhythmic phase, sleep homeostasis status, and other outputs. In some implementations, a plurality of biological samples is collected for measuring proteinaceous species, which can be useful for dynamic assessment or for combining assessments (e.g, arithmetic or statistical averaging of measurements and / or results). In some of these embodiments, analytes are measured with periodicity (e.g., hourly, every 4 hours, every 6 hours, every 8 hours, every 12 hours, daily, every two days, weekly, monthly, quarterly, yearly). Dependent on periodicity, various different assessments can be performed. For instance, multiple collections within a day can assess progression of rhythmic phases and / or sleep debt within a day. Multiple collections over days, weeks, or month can assess longitudinal changes of circadian clock and / or sleep homeostasis. In some implementations, when a longitudinal assessment is performed, collections are performed at or around the same time of day (e.g., within the same hour or less of the day each assessment).

[0088] In some embodiments, the method further comprises performing an assay to measure a set of proteinaceous species within a sample. The measurement assay can be performed directly on the collected biological sample or on a sample that is derived from biological sample. The appropriate preparation of the sample that will be utilized to measure proteinaceous species will depend on the assay that is performed.

[0089] Proteinaceous species can be measured by various different methodologies. Generally, proteinaceous species are assessed using an unspecific analysis technique or a specific detection technique. An example of an unspecific analysis technique is mass- spectrometry with liquid chromatography. Specific detection techniques can utilize a means to capture or otherwise identify particular species within a sample. Examples ofspecific detection techniques indude (but are not limited to enzyme-linked immunosorbent assay (ELISA), antibody arrays, DNA bar-coded antibodies (e.g., the Olink platform, Olink Holdings, Uppsala, Sweden) or aptamers (e.g., the Somalogic platform, Somalogic, Boulder, CO). In these techniques, an antibody or aptamer binds a particular proteinaceous species with high specificity to determine a relative amount of each particular proteinaceous species within the sample.

[0090] A biological sample can be collected from any individual that is to be assessed for rhythmic phase and / or sleep homeostasis status. In some instances, the individual is healthy or otherwise undiagnosed with the particular medical condition, in some instances, the individual has been diagnosed with a medical condition such as a sleep disorder, rhythm-related disorder, a cognitive or neurological disorder, a mental health condition, blindness, or other disorder that may be affected by rhythm or sleep homeostasis. In some instances, the individual is assessed for a condition related to a previously ascribed diagnosis (e.g., assessing rhythmic phase disruption of a diagnosed narcoleptic). In some instances, the individual has had recent behavior that affects rhythm or sleep (e.g., long-distant flights, working a night shift, caring for newborn child, etc.). Medical conditions, diagnoses, behavior assessments, and other related assessments are discussed in greater detail in subsequent sections.

[0091] Generally, a full panel of proteinaceous species can be assayed and measured, from just a few species to several thousand species, to all species identified in the sample. In several embodiments, the panel of proteinaceous species can be assayed and measured are circulating proteinaceous species. A number of proteinaceous species have been found to be useful in assessing rhythmic phase and sleep homeostasis / sleep debt. For examples of proteinaceous species that would be useful, see the description below and Appendix. In particular, examples of proteinaceous species useful for assessing circadian rhythm and diurnal rhythm are provided in Table 1A and 1 B, and Fig. 11. Within Supplemental Table A1 , proteinaceous species labeled as associated with circadian rhythm can be utilized to assess rhythmic phase. Within Table 2, proteinaceous species labeled as associated with sleep homeostasis status can be utilized to assess sleep homeostasis status and sleep deprivation.

[0092] Using the measurements of the proteinaceous species within a trained computational classifier, the sample can be classified (103) to belong to a rhythmic phase and / or sleep homeostasis status. Classification can be performed using any appropriate methodology. Generally, a computational classifier or regressor is trained to predict whether the sample was collected during a particular rhythmic phase or how much sleep debt that the individual is experiencing. Computational models that can be utilized include (but are not limited to) linear regression, polynomial regression, Cox proportional hazards regression, multiple linear regression, ridge regression, logistic regression, Lasso regression, stepwise regression, principal component analysis, Bayesian inference, elastic net, and random forest regression. In some embodiments, a classifier can utilize a regression technique to yield scaled results (e.g., a scale representative of rhythmic phases over a day, a scale representative of amount of sleep debt, or a likelihood of having a medical condition).

[0093] Any appropriate method for training a computational classifier for classifying a proteinaceous sample can be utilized. Generally, a cohort of individuals can be assessed for rhythmic phase or sleep homeostasis / sleep debt and have one or more biological samples collected. In many instances, the timing of biological sample collection is noted such that a rhythmic phase and / or sleep homeostasis is associated with the timing of collection. In many instances, recent sleep behavior (e.g., number of hours in each night for a period of time) is recorded such that a rhythmic phase and / or sleep homeostasis is associated with an amount sleep acquired. Various methods for assessing rhythmic phase and / or sleep homeostasis can be utilized. For example, rhythmic phase can be assessed via a DLMO test. One example of computing sleep debt is computing the number of hours that is needed for a week to feel rested and subtract the actual number of hours sleep obtained for the week. To train the computational classifier, each collected sample is associated with a rhythmic phase and / or an amount of sleep acquired / deprived. The proteinaceous species within each collected sample is measured. The resulting measurements and associated rhythmic phase and / or amount of sleep acquired / deprived are utilized to train the computational classifier.

[0094] As detailed in the example within the Examples section, several proteinaceous species have been found to be rhythmic and thus could be utilized as biomarkers ofrhythmic phases to be utilized as input into a classifier. It was found that training a lasso regression model with measurements of rhythmic proteinaceous species was able to yield a classifier capable of predicting the time the sample was collected in relationship to the individual’s DLMO. For instance, if a biological sample was collected at 11:00 AM and the individual’s DLMO was determined to be at 8:00 PM (or 20:00), the classifier was trained to predict that the sample was collected 9 hours before DLMO. Similarly, for a biological sample, a classifier can predict if a person is sleep deprived and / or how much sleep the person has missed over a recent period of time (e.g., within a day, within 2 days, within a week, etc.) or alternatively how much sleep the person has chronically missed over a longer period of time (e.g., over the past month). A computational model can also be trained to differentiate individuals having a sleep-wake disorder from individuals that have generally healthy sleep.

[0095] In some implementations, a classifier is trained to predict an individual’s DLMO based on the time of collection of the biological sample. In some implementations, a classifier is trained to predict an individual’s sleep homeostasis status. In some implementations, a model is trained to predict whether an individual has particular sleep disorder. Examples of sleep disorders that can be predicted are delayed sleep-wake phase disorder, chronic insomnia, REM sleep behavior disorder, narcolepsy. In some implementations, a model is further trained to predict whether an individual is a particular subtype of sleep disorder, such as subtypes of delayed sleep-wake phase disorder.

[0096] The number of proteinaceous species that are utilized in a classifier to predict a rhythmic phase and / or sleep homeostasis / sleep debt can vary. Generally, the greater number of proteinaceous species utilized, the better predictive power. In some implementations, measurements of proteinaceous species that provide the best predictive ability or utilized, which can be determined by the amount of predictive power that is provided. In various embodiments, measurements of at least two proteinaceous species are utilized within a classifier, measurements of at least three proteinaceous species are utilized within a classifier, measurements of at least four proteinaceous species are utilized within a classifier, measurements of at least five proteinaceous species are utilized within a classifier, measurements of at least ten proteinaceous species are utilized within a classifier, measurements of at least twenty proteinaceousspecies are utilized within a classifier, measurements of at least thirty proteinaceous species are utilized within a classifier, measurements of at least forty proteinaceous species are utilized within a classifier, measurements of at least fifty proteinaceous species are utilized within a classifier, or measurements of at least one hundred proteinaceous species are utilized within a classifier.

[0097] In some implementations, the computational model corrects for individualism. Generally, proteins and other proteinaceous species associated with rhythmic phase are cyclic, meaning the concentration of a species will increase up to a maximum during one circadian phase of the day and decrease to minimum during another phase. While the oscillatory nature of these species is generally conserved among all people, the concentration of some species can be different within each individual. Likewise, in the context of sleep homeostasis, the baseline concentration of some species can be different within each individual. Accordingly, in these implementations, a computational model can be utilized to adjust the input and / or output of the model to correct for individual variation. To do so, a computational model can be trained to learn a signature of individual using protein species within the sample, inclusive of non-cyclic circadian-related species and / or sleep-homeostasis-related species. A proteinaceous signature of an individual is to be understood to be a collection and concentration of proteinaceous species of the individual that are able to identify the individual. This will include species that deviate from the population norm. To train a computational model, proteinaceous species of a population are utilized as input to learn a population norm. An individual’s personal measurements of proteinaceous species is utilized as input into the model, providing a result indicating the deviation of the measurements from the population norm (e.g., statistically deviated beyond a threshold).

[0098] Various methodologies can be incorporated to correct for individualism via an identified individual proteinaceous species signature. In some implementations, the correction for individualism is performed prior to the classification of the sample to a rhythmic phase and / or sleep homeostasis. For example, a computational model can determine a signature of an individual, learning which circadian-rhythmic-related species and / or sleep-homeostasis-related species are deviated from the population norm. Deviated species can be removed from the input in the classifier for predicting rhythmicphase and / or steep homeostasis status, which will prevent their deviant effect on the prediction output. In some implementations, the correction for individualism is performed after the classification of the sample to a rhythmic phase and / or steep homeostasis. For example, a computational model can determine a signature of an individual, learning which circadian-rhythmic-related species and / or sleep-homeostasis-related species are deviated from the population norm, and then normalize the classification result by adjusting for the result based on the deviated species. The classification model operates with presumption that the concentration of proteinaceous species used as input to predict rhythmic phase and steep homeostasis status are centered at the population norm, but generally each individual will have their own center based on their signature. Using an individual’s proteinaceous species signature, the individual’s personal center is computed and then utilized to adjust the results of the classification model based on the difference between the center of the population norm and the individual’s center. It should be understood that "center can be a baseline, a minimum, a maximum, a mean, a median, or any other relevant point of species concentration, especially of species that oscillate.

[0099] A model for correcting for individualism can be incorporated into the computational framework comprising the classification model by any appropriate method. In various implementations, a model for correcting for individualism is directly linked before, after, and / or before and after the classification model, yielding a single application that performs the classification with correction for individualism. Alternatively, each model can be maintained individually and can be modular, where each model is performed as desired by the user or by some determinant (e.g., statistical deviation computation). It should also be understood that correction for individualism can be performed the classification, after the classification, or both before and after the classification. In some implementations, a method can comprise a correction to remove deviated species from the input into the classifier, then perform the classification, and further adjust the classification result based on the individual’s signature center.

[0100] Having classified a sample with a certain rhythmic phase and / or a sleep debt status, a further diagnostic assessment or a treatment can optionally be performed (105) on the individual. The results of the classification of the biological sample can be utilized in a number of different manners. For instance, the timing of an individual’s DLMO canbe determined and thus an assessment can be made on whether the individual’s DLMO should be adjusted to improve sleep, tiredness, or any other symptom or condition related to circadian rhythm. Likewise, an individual can be classified as having too little sleep or too much sleep and thus an assessment can be made on whether the individual should adjust the amount of sleep they are having. A number of medical conditions are affected by the timing of an individual’s DLMO or sleep amount, such as (for example) narcolepsy, chronic insomnia, hypersomnia, advanced sleep-wake phase disorder, delayed sleepwake phase disorder, irregular sleep-wake rhythm disorder, jet lag disorder, non-24-hour sleep-wake rhythm disorder, and shift work disorder. Accordingly, classification of the biological sample can be utilized in diagnostic assessment of DLMO, sleep homeostasis, or other medical conditions affected by rhythm or sleep. Thus, in some instances, a classification of an individual’s biological sample of proteinaceous species is utilized to inform whether a medical intervention is to be performed on the individual. In some instances, a classification result suggests that a further diagnostic procedure should be performed to determine whether the individual has particular medical condition. In some instances, a classification result suggests that the individual would benefit from administration of a treatment. For instance, melatonin and / or light therapy can be administered to adjust an individual’s DLMO. Likewise, a hypnotic agent can be utilized to induce sleep in an individual having too little sleep or a stimulant can be utilized to induce wake to an individual having too much sleep.

[0101] In one example, an individual presenting symptoms of delayed sleep-wake phase disorder can have their DLMO assessed to determine whether a treatment to adjust circadian cycle would be of benefit. Two subtypes of delayed sleep-wake phase disorder are recognized: (1 ) circadian delayed and (2) non-circadian (see J. M. Murray, et al., Sleep. 2017 Jan 1 ;40(1 ), the disclosure of which is hereby incorporated by reference). The two subtypes can be diagnosed by timing of DLMO and response to the melatonin onset. If melatonin onset is delayed upon dimming of light, then the individual has a circadian delayed sleep-wake phase disorder subtype. These individuals have a delayed circadian phase but are responsive to melatonin and generally fall asleep within two to three hours of melatonin release. Accordingly, individuals with circadian delayed sleep-wake phase disorder subtype have a normal response to melatonin but improperdelay of release of melatonin upon light dimming. Conversely, if an individual releases melatonin in response to dimming of light and prior to desired sleep onset, but maintains wakefulness beyond two to three hours after DLMO, then the individual has non-circadian delayed sleep-wake phase disorder subtype. Melatonin receptor agonists are a beneficial treatment for the circadian misaligned subtype but not for the circadian aligned subtype. These individuals have a an on-time circadian phase but are nonresponsive to melatonin resulting in delayed sleep onset. Accordingly, individuals with non-circadian delayed sleep-wake phase disorder subtype have a normal release of melatonin upon light dimming but improper response to melatonin that results in delay of sleep. Because the two subtypes can be differentiated by DLMO timing, a computational model for determining DLMO can be utilized to predict the subtypes of individual diagnosed with delayed sleep-wake phase disorder. Individuals with circadian delayed sleep-wake phase disorder subtype would benefit from a treatment regimen that includes administration of a melatonin receptor agonist prior to desired sleep onset, which can also help reset their circadian phase. Individuals with circadian delayed sleep-wake phase disorder subtype would benefit from a melatonin receptor agonist. Individuals of the non-circadian subtype can be administered hypnotic agents, such as benzodiazepines, non-benzodiazepine sedative-hypnotics (e.g., Z-type drugs), orexin receptor antagonists, sedative antidepressants, or antihistamine.

[0102] Several biological processes are affected by rhythmic phase and / or sleep homeostasis / sleep debt, which also affects timing of pharmaceutical intervention (see, e.g., M. D. Ruben, et al., Science. 2019 Aug 9;365(6453):547-549; and W. Ruan, et al., Nat Rev Drug Discov. 2021 Apr;20(4):287-307; the disclosures of which are hereby incorporated by reference). One example is cell division and it has been found that immune checkpoint inhibitors (ICIs) utilized in treatments in cancer can have improved efficacy dependent on timing of administration. It has been demonstrated that early morning infusions of ICI had greater efficacy than later infusions, (see S. Catozzi, et al., Eur J Cancer. 2024 Mar;199:113571 , the disclosure of which is hereby incorporated by reference). Accordingly, there is benefit in determining an individual’s DLMO in order to improve efficacy of various treatments by timing administration of such treatments based on the individual’s DLMO. This can be especially useful for individuals that have a DLMOthat deviates from the norm, such as individuals with sleep disorder, have impaired vision or blindness, work night shifts, experience jetlag, or have irregular sleep schedules. Thus, in some instances, a classification of an individual’s biological sample of proteinaceous species is utilized to improve treatment regimens by an administration of the treatment in accordance with a determined DLMO. In some implementations, when an individual is determined to have an unaligned circadian clock, the individual is administered an ICI at the time that would be equivalent to early morning for based on their circadian clock. In some implementations, when an individual is determined to have an unaligned circadian clock, the individual is administered melatonin receptor agonist and / or light therapy in addition to the ICI treatment, where the melatonin and / or light therapy is to realign the circadian clock. The melatonin receptor agonist and / or light therapy can be administered ahead of initiating the course of ICI treatment and / or during the course of ICI treatment. For more on checkpoint inhibitors and the relationship to timing, see, e.g., M. Nomura, et al., Esophagus. 2023, 20(4)722-731 ; A. Rousseau, et al., Eur J Cancer. 2023,182:107- 114; L. Goncalves, et al., Cells. 2023, 12(16):2068; N. Dizman, et al., Clin Genitourin Cancer. 2023, 21 (5):530-536; A. Karaboue, et al., Cancers (Basel). 2022, 14(4):89; A. Karaboue, et al., Br J Cancer. 2024 Sep; 131 (5)783-796; and D. C. Qian, et al., Lancet Oncol. 2021 , 22(12): 1777-1786; the disclosures of which are each incorporated herein by reference.

[0103] Method 100 can further optionally repeat (107) obtaining measurements of a set of proteinaceous species of a sample and classifying that sample with a rhythmic phase or sleep homeostasis status. By repeating the measurement of proteinaceous species and classification, an individual can have their rhythms and / or sleep homeostasis monitored. Monitoring can be performed for various reasons. For example, an individual can be monitored to determine whether a treatment regimen is improving a rhythmic and / or sleep homeostasis related medical condition. In another example, when efficacy of a treatment is affected by timing of administration, the individual’s DLMO can be monitored to adjust the timing of administration such that high efficacy of the treatment is maintained.

[0104] While specific examples of classifying an individual’s sample with a rhythmic phase or sleep homeostasis / sleep debt status are described above, one of ordinary skillin the art can appreciate that various steps of the process can be performed in different orders and that certain steps may be optional according to some embodiments of the invention. As such, it should be clear that the various steps of the process could be used as appropriate to the requirements of specific applications. Furthermore, any of a variety of processes for classifying an individual’s sample with a rhythmic phase or sleep homeostasis / sleep debt status appropriate to the requirements of a given application can be utilized in accordance with various embodiments of the disclosure.Proteinaceous Biomarkers for Diagnostic Use

[0105] Systems and methods of the disclosure can utilize a panel of proteinaceous species biomarkers to assess rhythmic phase, sleep homeostasis, and related medical conditions. Examples of biomarkers that can be utilized are provided throughout the Examples section, and specifically within Tables 1A, 1 B, and 2, and Fig. 11. In particular, the proteinaceous species within Table 1A labeled as associated with circadian rhythm can be utilized to assess rhythmic phase. Likewise, the proteinaceous species within Table 2 can be utilized to assess sleep homeostasis / sleep debt.

[0106] Examples of proteinaceous species that can be measured for assessment of rhythmic phase include (but are not limited to) angiopoietin-related protein 1 , cathepsin F, pro-opiomelanocortin, prolactin, metalloproteinase inhibitor 4, hyaluronidase-1 , thyroid stimulating hormone, kallikrein-7, secretogranin-1 , cysteine-rich secretory protein LCCL domain-containing 2, hemopexin, zymogen granule protein 16 homolog B, pancreatic hormone, follistatin-related protein 3, neutrophil cytosol factor 1 , heparan-sulfate 6-0- sulfotransferase 2, cGMP-dependent 3',5'-cyclic phosphodiesterase, HLA class I alpha chain G, lactotransferrin, proprotein convertase subtilisin / kexin type 9, secreted and transmembrane protein 1 , desmocollin-2, succinate dehydrogenase assembly factor 2- mitochondrial, and lysosomal protective protein.Detecting and Measuring Levels of Biomarkers

[0107] Proteinaceous species biomarkers in a biological sample (e g., blood, plasma, stool, urine, or biopsy) can be assessed by a number of suitable methods. Generally, methods can be broken into specific detection methods and unspecific detectionmethods. Specific detection methods rely on a panel of molecules for specifically binding to a proteinaceous species to form a complex and then detecting such binding. Unspecific methods generally run a proteinaceous sample through a detector and assessing the results to determine which proteinaceous species were present. Suitable methods for specific detection of proteinaceous species include (but are not limited to) EUSA, antibody arrays, DNA bar-coded antibodies, and aptamers. Suitable methods for unspecific detection of proteinaceous species include chromatography (e.g., high- performance liquid chromatography (HPLC), gas chromatography (GC), liquid chromatography (LC)), mass spectrometry (e.g., MS, MS-MS), NMR, electrophoresis, and combinations thereof. For example, mass spectrometry can be combined with chromatographic methods, such as liquid chromatography (LC), gas chromatography (GC), or electrophoresis to separate the metabolite being measured from other components in the biological sample. See, e.g., Hyotylainen (2012) Expert Rev. Mol. Diagn. 12(5):527-538; Beckoned et al. (2007) Nat. Protoc. 2(11 ):2692-2703; O’Connell (2012) Bioanalysis 4(4):431 -451 ; and Eckhart et al. (2012) Clin. Transl. Sci. 5(3):285-288; the disclosures of which are herein incorporated by reference.

[0108] ELISA is a technique in which antibodies are bound to a substrate or bead. A sample of proteinaceous species is mixed with the antibodies to form a complex. Colorimetric or other signal-generating techniques are utilized to detect the complex.

[0109] Antibody arrays utilize a panel of different antigen-recognizing antibodies clustered in spots and bound to a membrane or other substrate, forming an array-like pattern of spots. A sample of proteinaceous species is mixed with the antibodies to form a complex with immobilized antibody. After washing off the excess sample, set of antibodies for each antigen assessed is added on top of the membrane, where each of these antibodies includes a conjugated signal-generating molecule, such as a peroxidase enzyme. The presence and relative concentration of each proteinaceous species assessed can be determined by assessing the location within the array comprising the spot of immobilized antibody for detecting the species.

[0110] The DNA bar-coded antibody-based technology utilizes a panel antibodies that can specifically recognize a set of proteinaceous species. Each unique antibody includes a particular DNA bar code attached thereon. The antibodies are utilized to bindproteinaceous species to form complex within the sample. Upon clearance of unbound complexes, the DNA bar codes can be detected utilizing a sequencing technique, thus providing detection of presence and relative amount of particular proteinaceous species in the sample.

[0111] The aptamer-based technology utilizes a panel aptamers that can specifically recognize a set of proteinaceous species. Each aptamer sequence is designed to uniquely bind a particular proteinaceous species. The aptamers are utilized to bind proteinaceous species to form complex within the sample. Upon clearance of unbound complexes, the aptamer sequences can be detected utilizing a sequencing technique, thus providing detection of presence and relative amount of particular proteinaceous species in the sample.

[0112] In various embodiments, proteinaceous species biomarkers in a sample can be separated by high-resolution electrophoresis, e.g., one or two-dimensional gel electrophoresis. A fraction containing a biomarker can be isolated and further analyzed by gas phase ion spectrometry. Preferably, two-dimensional gel electrophoresis is used to generate a two-dimensional array of spots for the biomarkers. See, e.g., Jungblut and Thiede, Mass Spectr. Rev. 16:145-162 (1997).

[0113] Two-dimensional gel electrophoresis can be performed using any one of various methods known in the art. See, e.g., Deutscher ed., Methods In Enzymology vol. 182. Typically, lipid biomarkers in a sample are separated by, e.g., isoelectric focusing, during which biomarkers in a sample are separated in a pH gradient until they reach a spot where their net charge is zero ( / .e., isoelectric point). This first separation step results in one-dimensional array of biomarkers. The biomarkers in the one-dimensional array are further separated using a technique generally distinct from that used in the first separation step. For example, in the second dimension, biomarkers separated by isoelectric focusing are further resolved using a polyacrylamide gel by electrophoresis.

[0114] Biomarkers in the two-dimensional array can be detected using any suitable methods known in the art. For example, proteinaceous species biomarkers in a gel can be labeled or stained (e.g., Nile red). If gel electrophoresis generates spots that correspond to the molecular weight of one or more biomarkers, the spot can be further analyzed by densitometric analysis or gas phase ion spectrometry. For example, spotscan be excised from the gel and analyzed by gas phase ion spectrometry. Alternatively, the gel containing biomarkers can be transferred to an inert membrane by applying an electric field. Then a spot on the membrane that approximately corresponds to the molecular weight of a biomarker can be analyzed by gas phase ion spectrometry. In gas phase ion spectrometry, the spots can be analyzed using any suitable techniques, such as MALDI or SELDI.

[0115] High performance liquid chromatography (HPLC) can be used to separate a mixture of biomarkers in a sample based on their different physical properties, such as polarity, charge and size. HPLC instruments typically consist of a reservoir, the mobile phase, a pump, an injector, a separation column, and a detector. Biomarkers in a sample are separated by injecting an aliquot of the sample onto the column. Different biomarkers in the mixture pass through the column at different rates due to differences in their partitioning behavior between the mobile liquid phase and the stationary phase. A fraction that corresponds to the molecular weight and / or physical properties of one or more biomarkers can be collected. The fraction can then be analyzed by gas phase ion spectrometry to detect biomarkers.

[0116] Mass spectrometry is useful for detection of proteinaceous species biomarkers. On example is laser desorption time-of-f light mass spectrometer in which a substrate or a probe comprising biomarkers is introduced into an inlet system. The biomarkers are desorbed and ionized into the gas phase by laser from the ionization source. The ions generated are collected by an ion optic assembly, and then in a time-of-flight mass analyzer, ions are accelerated through a short high voltage field and let drift into a high vacuum chamber. At the far end of the high vacuum chamber, the accelerated ions strike a sensitive detector surface at a different time. Since the time-of-flight is a function of the mass of the ions, the elapsed time between ion formation and ion detector impact can be used to identify the presence or absence of markers of specific mass to charge ratio.

[0117] Biomarkers on the substrate surface can be desorbed and ionized using gas phase ion spectrometry. Any suitable gas phase ion spectrometer can be used as long as it allows biomarkers on the substrate to be resolved. Preferably, gas phase ion spectrometers allow quantitation of biomarkers. A gas phase ion spectrometer can be a mass spectrometer. In a typical mass spectrometer, a substrate or a probe comprisingbiomarkers on its surface is introduced into an iniet system of the mass spectrometer. The biomarkers are then desorbed by a desorption source such as a laser, fast atom bombardment, high energy plasma, electrospray ionization, thermospray ionization, liquid secondary ion MS, field desorption, etc. The generated desorbed, volatilized species consist of preformed ions or neutrals which are ionized as a direct consequence of the desorption event. Generated ions are collected by an ion optic assembly, and then a mass analyzer disperses and analyzes the passing ions. The ions exiting the mass analyzer are detected by a detector. The detector then translates information of the detected ions into mass-to-charge ratios. Detection of the presence of biomarkers or other substances will typically involve detection of signal intensity. This, in turn, can reflect the quantity and character of biomarkers bound to the substrate. Any of the components of a mass spectrometer (e.g., a desorption source, a mass analyzer, a detector, etc.) can be combined with other suitable components described herein or others known in the art.Diagnostics, dedications and Dietary Supplements

[0118] Various embodiments are directed to assessment and / or amelioration of ailments as related to rhythmic phase and / or sleep homeostasis, including diagnostic methods, treatments, and / or dietary supplementation. As described in the examples herein, a panel of proteinaceous species can be utilized within a computational model to determine an individual’s rhythmic phases and / or sleep homeostasis status, which can be utilized to inform various treatment options.

[0119] Many embodiments are directed towards diagnostics for determining an individual’s rhythmic phase. In some embodiments, an individual’s DLMO is predicted by assessing proteinaceous species of one or more biological samples of the individual using a computational classifier. In some instances, upon determining an individual’s DLMO, the individual can be diagnosed as needing an adjustment in phase timing. In some instances, upon determining an individual’s DLMO, an individual is prescribed a treatment regimen that comprises administration of one or more drugs, light therapy, and / or behavioral therapy.

[0120] An example of a diagnostic method and treatment for adjusting phase timing is as follows:« Receive measurements of proteinaceous species of a biological sample of an individual« Enter the measurements of the proteinaceous species into a computational classifier to determine 1 ) within which rhythmic phase that the biological sample was collected and 2) the amount of sleep deprivation at the time of collection» Administer behavioral therapy, a modulator of circadian rhythm(e.g., melatonin receptor agonists), or light therapy at specific times to adjust the timing of rhythmic phases of the individualIn some embodiments, the method further comprises measuring the amount of each proteinaceous species of a panel. In some embodiments, the method further comprises collecting a biological sample of the individual.

[0121] In some embodiments, it is determined that an individual’s DLMO is delayed and the individual is administered a melatonin receptor agonist. The melatonin receptor agonist can be taken at a time that resets the DLMO to an earlier timepoint, such as twenty minutes to an hour before the desired time to sleep. Examples of melatonin receptor agonists include melatonin, tasimelteon, ramelteon, and agomelatine. Alternatively, or in addition to the use of melatonin receptor agonists, when DLMO is delayed, light therapy can be administered in the morning at a time that resets wakefulness to an earlier timepoint. In some embodiments, when DLMO is delayed, a hypnotic compound is administered at night to promote sleep (e.g. benzodiazepines, non-benzodiazepine sedative-hypnotics, orexin receptor antagonists, sedative antidepressants, antihistamine).

[0122] In some embodiments, it is determined that an individual’s DLMO is advanced and the individual is administered light therapy at a time that resets the DLMO to an earlier timepoint. One therapy regimen can include performing light therapy at the time melatonin onset is occurring such that melatonin release is pushed back. An alternative therapy for advanced DLMO is to administer a melatonin receptor agonist at core body temperature nadir (i.e., the lowest body temperature during sleep). In some embodiments, when DLMO is advanced, a stimulant is administered (e.g., caffeine, wake promoting compounds such as dopaminergic stimulants, amphetamines, H3 antagonists, orexin agonists) during the day to delay onset of DLMO.

[0123] Many embodiments are directed towards diagnostics for assessment of delayed sleep-wake phase disorder subtype. Two subtypes of delayed sleep-wake phase disorder subtype that are recognized include circadian delayed and non-circadian. When an individual is diagnosed as having delayed sleep-wake phase disorder, determination of subtype can inform which treatment options would be beneficial. Subtype can be determined by assessing an individual’s DLMO, which can be predicted by assessing proteinaceous species of one or more biological samples of the individual using a computational classifier. When the classifier predicts that DLMO is delayed, the individual is diagnosed as having the circadian delayed subtype and would benefit from melatonin receptor agonists and / or light therapy. When the classifier predicts that DLMO is not delayed, the individual is diagnosed as having non-circadian subtype and would not benefit from melatonin receptor agonists. Individuals diagnosed as having non-circadian subtype can be administered a hypnotic agent, such as benzodiazepines, nonbenzodiazepine sedative-hypnotics (e.g., Z-type drugs), orexin receptor antagonists, sedative antidepressants, or antihistamine.

[0124] An example of a diagnostic method for delayed sleep-wake phase disorder subtype and treatment is as follows:• Determine that an individual has a diagnosis of delayed sleep-wake phase disorder• Receive measurements of proteinaceous species of a biological sample of the individual• Enter the measurements of the proteinaceous species into a computational classifier to determine the individual's DLMO. If DLMO is delayed, administer melatonin receptor agonists prior to desired time of sleep and / or light therapy in the morning. If DLMO is not delayed, do not administer melatonin receptor agonists or light therapyIn some embodiments, the method further comprises measuring the amount of each proteinaceous species of a panel. In some embodiments, the method further comprises collecting a biological sample of the individual.

[0125] In some embodiments, it is determined that an individual’s DLMO is delayed and has circadian delayed sleep-wake phase disorder subtype and the individual isadministered a melatonin receptor agonist. The melatonin receptor agonist can be taken at a time that resets the DLMO to an earlier timepoint, such as twenty minutes to an hour before the desired time to sleep. Examples of melatonin receptor agonists include melatonin, tasimelteon, ramelteon, and agomelatine. Alternatively, or in addition to the use of melatonin receptor agonists, when DLMO is delayed, light therapy can be administered in the morning at a time that resets wakefulness to an earlier timepoint.

[0126] In some embodiments, it is determined that an individual’s DLMO is not delayed and the individual has non-circadian delayed sleep-wake phase disorder. The individual would not benefit from melatonin receptor agonists in the evening and / or light therapy in the morning. In some embodiments, individuals diagnosed as having non-circadian subtype can be administered a hypnotic agent, such as benzodiazepines, nonbenzodiazepine sedative-hypnotics (e.g., Z-type drugs), orexin receptor antagonists, sedative antidepressants, or antihistamine prior to a desired time to sleep.

[0127] Many embodiments are directed towards diagnostics for determining an individual’s rhythmic phase for the purpose of improving efficacy of a treatment that is affected by circadian phase. In some embodiments, an individual's DLMO is predicted by assessing proteinaceous species of one or more biological samples of the individual using a computational classifier. In some instances, upon determining an individual’s DLMO and timing of circadian phases, the individual can be administered a drug at a time of improved efficacy in accordance with their circadian clock. In some instances, upon determining that an individual’s DLMO and timing of circadian phases is either delayed or advanced, the individual is administered a treatment to reset and align the circadian clock.

[0128] Numerous treatments for various ailments can be optimized based on circadian clock. Examples medications shown to be affected by timing of administration include (but are not limited to) statins for high cholesterol (e.g., simvastatin), alpha-blockers for hypertension (e.g., phenatolamine), beta-blockers for hypertension (e.g., propanalol, nebivolol), angiotensin II receptor blockers for hypertension (e.g., valsartan, telmisartan, olmesartan, candesartan), chemotherapeutics for cancer (e.g., L-OHR, 5-FU, FA, DDC, DOC), immune checkpoint inhibitors for cancer (e.g., nivolumab, pembrolizumab, dostarlimab, sintilimab cemiplimab, ipilimumab, tremelimumab, tezolizumab, durvalumab, avelumab), antihistamines for allergic dermatitis or rhinitis (e.g., cyproheptadine,terfenadine), NSAIDs for arthritis (e.g., flurbiprofen, indomethacin), corticosteroids for arthritis (e.g., prednisone), antimetabolite for arthritis (e.g., methotrexate), phosphodiesterase inhibitors for asthma (e.g., theophylline), p2y12 inhibitors for clotting (e.g., clopidogrel), Factor Xa inhibitors for anticoagulation (e.g., rivaroxaban) and proton pump inhibitors for gastroesophageal reflux disease (e.g., omeprazole, rabeprazole) (see M. D. Ruben, et al., Science. 2019 Aug 9; 365(6453): 547-549, inclusive of supplemental material; the disclosure of which is hereby incorporated by reference). It is noted that drugs that are fast acting and / or have a short half-life are more likely affected by circadian phase and as such optimal timing of these drugs may be more imperative.

[0129] An example of a diagnostic method for optimizing a treatment regimen is as follows:• Prescribe an individual with a treatment for a medical condition in which the treatment efficacy can be improved by timing administration in reference to a rhythmic phase• Receive measurements of proteinaceous species of a biological sample of the individual• Enter the measurements of the proteinaceous species into a computational classifier to determine the individual’s DLMO and timing of circadian phases• Administer the treatment at a time that improves efficacy in accordance with the individual’s rhythmic phases and sleep deprivation status• Optional: Administer behavioral therapy, a modulator of circadian rhythm(e.g., melatonin receptor agonists), or light therapy at specific times to reset the timing of rhythmic phases of the individual, which can be performed prior to and / or treatment with drugs affected by circadian phaseIn some embodiments, the method further comprises measuring the amount of each proteinaceous species of a panel. In some embodiments, the method further composes collecting a biological sample of the individual.

[0130] Several embodiments are directed to the use of medications and / or dietary supplements to treat an individual to achieve a beneficial result. In some embodiments, medications and / or dietary supplements are administered in a therapeutically effective amount as part of a course of treatment. As used in this context, to "treat" means toameliorate at least one symptom of the disorder to be treated or to provide a beneficial physiological effect. For example, one such amelioration of a symptom could improve sleep and / or tiredness. A therapeutically effective amount can be an amount sufficient to prevent, reduce, ameliorate or eliminate the symptoms of diseases or medical conditions susceptible to such treatment, such as, for example, improving rhythmicity or improving efficacy of treatment based on an individual’s rhythmic phases.

[0131] In various implementations, one or more compounds (e.g., melatonin) are administered to an individual in the form of a therapeutic and / or dietary supplement. The one or more compounds described herein may be combined with any number of ingredients that facilitate effective ingestion, administration, delivery, adsorption, distribution, and metabolism of the active agents.

[0132] The therapeutic compositions may be formulated for efficient delivery by a selected route. The therapeutic compositions may be formulated in combination with pharmaceutically acceptable excipients, carriers, diluents, release formulations and other drug delivery or drug targeting vehicles.

[0133] The therapeutic compositions may be formulated for administration to a subject. Administration, as used herein, may encompass the act of providing the agent to the subject, introducing the agent into the subject, and may include self-administration. The therapeutic compositions can be formulated for oral delivery. Oral formulations may include any number of fillers, flavoring agents, coloring agents, sweeteners, and other compositions used in the art for oral delivery of agents. In some implementations, therapeutics and / or dietary supplements are ingested by a subject, which can be performed by any method of ingestion. For example, the one or more lipid compositions may be provided as capsules, tablets, foods, and / or liquids for administration. Further, the one or more lipid compositions can be provided as an oil or emulsified in an aqueous solution (e.g., in the form of lipid micelles).

[0134] In some implementations, therapeutic compositions are formulated for administration via non-oral routes, including, for example, intravenous, intra-arterial, intraperitoneal, intrapulmonary, oral, inhalation, intravesicular, intramuscular, intratracheal, subcutaneous, intraocular, intrathecal, transmucosa!, and transdermal delivery.

[0135] In some implementations, therapeutic are formulated and administered as a dietary supplement, A dietary supplement is a composition that is administered orally, for example, in food or drink, with food or drink, or which otherwise supplements the regular intake of food. From a regulatory standpoint, a dietary supplement is generally a composition that can be purchased and used without a prescription. For example, in the United States, per the Food and Drug Administration and The Dietary Supplement Health and Education Act of 1994, a dietary supplement is “a product intended to supplement the diet that bears or contains one or more of the following dietary ingredients: a vitamin, a mineral, an herb or other botanical, an amino acid, a dietary substance for use by man to supplement the diet by increasing the total dietary intake, or a concentrate, metabolite, constituent, extract, or combination of any ingredient just described.” Dietary supplements may encompass any formulation for oral delivery, as described herein.

[0136] Examples of rhythm ic / sleep related medical conditions that can be treated include (but are not limited to) narcolepsy, chronic insomnia, hypersomnia, advanced sleep-wake phase disorder, delayed sleep-wake phase disorder, irregular sleep-wake rhythm disorder, jet lag disorder, non-24-hour sleep-wake rhythm disorder, and shift work disorder. Various treatments include behavioral therapy, phase resetting agents, hypnotic compounds, stimulants, and light therapy.

[0137] Examples of medications with efficacy that is affected by timing of administration include (but are not limited to) statins, alpha-blockers, beta-blockers, angiotensin II receptor blockers, chemotherapeutics, immune checkpoint inhibitors, antihistamines, NSAIDs, corticosteroids, phosphodiesterase, p2y12 inhibitors, and proton pump inhibitors.Computational Processing System

[0138] A computational processing system to predict circadian phase and sleep homeostasis status in accordance with various embodiments of the disclosure typically utilizes a processing system including one or more of a CPU, GPU and / or other processing engine. In some embodiments, the computational processing system is housed within a computing device. In certain embodiments, the computational processing system is implemented as a software application on a computing device such as (but not limited to) mobile phone, a tablet computer, and / or portable computer.[0i 39] A computational processing system in accordance with various embodiments of the disclosure is illustrated in Fig. 2. The computational processing system 200 includes a processor system 202, an I / O interface 204, and a memory system 206. As can readily be appreciated, the processor system 202, I / O interface 204, and memory system 206 can be implemented using any of a variety of components appropriate to the requirements of specific applications including (but not limited to) CPUs, GPUs, ISPs, DSPs, wireless modems (e.g., WiFi, cellular, Bluetooth modems), serial interfaces, volatile memory (e.g., DRAM) and / or non-volatile memory (e.g., SRAM, and / or NAND Flash). The memory system can be configured to store various applications and computational models, each of which are optional and / or combinable in any fashion. Applications include DLMO prediction application 208, sleep homeostasis status prediction application 210, and personalized protein signature correction application 212. Each listed applications can comprise a machine-learing model and utilize proteinaceous species measurements as input, which can be entered via the I / O interface or gathered via a connection (e.g., WiFi, cellular Bluetooth). When executed, the various applications and models are each capable of configuring the processing system to implement computational processes including (but not limited to) the computational processes described above and / or combinations and / or modified versions of the computational processes described above.

[0140] While specific computational processing systems are described above with reference to Fig. 2, it should be readily appreciated that computational processes and / or other processes utilized for prediction of circadian phase and / or sleep homeostasis status in accordance with the various embodiments of the disclosure can be implemented on any of a variety of processing devices including combinations of processing devices. Accordingly, computational devices in accordance with embodiments of the disclosure should be understood as not limited to specific computational processing systems. Computational devices can be implemented using any of the combinations of systems described herein and / or modified versions of the systems described herein to perform the processes, combinations of processes, and / or modified versions of the processes described herein.EXAMPLES

[0141] Bioinformatic and biological data support the systems and methods of assessing rhythmic proteinaceous species and applications thereof. Examples of methods and applications related to prediction of circadian rhythmic phase and sleep homeostasis status are provided. Additionally, examples proteinaceous species biomarkers that can be utilized within computational models to perform prediction and diagnostics are described. Data results and analysis are provided. The data further provides support the various therapeutics and dietary supplements that can be utilized as treatment.CIRCADIAN PROTEIN EXPRESSION PATTERNS IN HEALTHY YOUNG ADULTS

[0142] Circadian rhythms are powerful regulators of physiology. In mammals, these are driven by a central pacemaker located in the suprachiasmatic nucleus (SCN), which receives direct input from the visual system, allowing synchronization to the external 24- hour day. The central clock, in turn, is connected through neuronal and endocrine networks to other tissues which have their own endogenous circadian clocks, allowing each system to coordinate temporal programs optimal to the organism. Although much is known regarding how each cell generates its own circadian rhythm through a process involving transcriptional / translational feedback loops, the synchronization between peripheral and central clocks is poorly understood, and there is evidence these systems can become dissociated. As an example, in rodents and humans, restricting food availability can shift the phase of metabolic processes to times when food is available, with deleterious metabolic consequences. For this reason, there is interest in understanding and describing how these processes are organized.

[0143] Traditionally, timing (or phase) of the human central circadian clock is measured using the Dim Light Melatonin Onset (DLMO), because secretion of melatonin by the pineal gland is controlled via a multi-synaptic pathway from the SCN through the superior cervical ganglion. Melatonin is at very low or non-detectable levels in blood and saliva during the daytime and begins secretion in the evening, reaching peak levels in the middle of the subjective night. While the rhythm of melatonin has served a useful purpose for assessing the phase of the central clock for research purposes, it does have severaldrawbacks. First, melatonin is acutely suppressed by light, and assessment of melatonin in saliva or blood requires the participant to remain in dim light throughout the sampling segment. Second, measurement from saliva requires an awake participant, while measurement from blood requires either multiple needle sticks (again waking the participant) or a special blood sampling system that allows sampling without waking the participant. These issues, together with the duration of melatonin sampling required to obtain a phase estimate have severely limited the assessment of melatonin timing in sleep disorders patients. Furthermore, while melatonin provides useful information about central circadian phase, it does not provide any indication of the phase of peripheral oscillators, nor whether peripheral rhythms are synchronized to the central clock or to each other.

[0144] For these reasons, researchers have sought alternatives for assessing endogenous circadian phase in humans. In animals, tissues can be collected and analyzed for metabolites, gene expression, or protein content to assess the phase of both the central and peripheral oscillators. Such studies have led to a wealth of information on how organ clocks are organized, with the limitations that most animal models are nocturnal, inbred, and live in artificial lighting and environmental conditions. Animal models have shown that by measuring multiple parameters at a given time point and considering a stable relationship between phases of various processes, it is possible to derive the central circadian phase with a single time point, a concept called a “time stamp". Subsequently, attempts have been made at establishing time stamp measures that could replace melatonin phase in humans, but in most cases to achieve accuracy within 1-2 hours of DLMO phase at least two samples collected hours apart have been needed. One exception is BodyTime, a measure of circadian phase from gene expression in isolated monocytes derived from a single blood sample, which has shown good performance in studies against DLMO (N. Wittenbrink, et al., J Clin Invest. 2018 Aug 31 ;128(9):3826-3839, the disclosure of which is hereby incorporated by reference). Although powerful, BodyTime measures the circadian clock in a specific immune tissue that could be perturbed in conditions such as infections. Nonetheless, it has the advantage of measuring core circadian genes themselves and thus has heuristic value.

[0145] In the present work, we hypothesized that measuring the circulating plasma proteome might lead to future development of a single sample proteomic circadian phase measure, and herein describe proteins that have circadian influences. Indeed, it is now possible to measure an increasing number of proteins in smaller and smaller amounts of blood, either using Olink, a DNA bar-coded antibody-based technology, or Somalogic, an aptamer-based technology (G. S. Omenn, J Proteome Res. 2023 Apr 7;22(4): 1024-1042, the disclosure of which is hereby incorporated by reference). The results of Olink and Somalogic are comparable in relation to mass spectrometry. The reliability of these technologies is revealed by many genetic effects that have been shown to regulate levels of individual proteins and that are mostly located close to the gene encoding each analyte (Cis-QTL). These signals are often themselves also known as expression QTLs. These proteins are secreted or the result of normal tissue shedding and damage and can be both of intra and extracellular origin. Proteomics offers several advantages over metabolomics because many metabolites are unstable and of gastrointestinal origin, as weli as advantages over transcriptom ics, which utilizes whole blood where different cell types show rhythmic fluctuations in number and internal gene expression. Furthermore, proteins are closer to physiology than gene transcripts, as they are the structural and acting component of the human body. Although transcript and protein amounts correlate within tissues, there are many exceptions. Another advantage of measuring the plasma proteome is that it is likely to contain many secreted proteins and hormones that are used in inter-organ communications.

[0146] Here, we carried out a study to explore whether the blood proteome shows 24- hour rhythmic activity and whether any such rhythmic proteins show endogenous circadian variation. Somalogic currently offers a larger number of analytes (-7,000 proteins) from single small blood samples (200 pL plasma) and is therefore the platform we elected to use.MethodsParticipants

[0147] Samples were collected from two studies carried out at the Intensive Physiological Monitoring Unit of the Center for Clinical Investigation (CCI) at Brigham andWomen’s Hospital (BWH), part of the Harvard Catalyst Clinical and Translational Science Center.

[0148] In the first study, 8 adults (2 men, 6 women) between ages 21 and 35 were recruited for a 5-day study. To be included, participants had to have a self-reported habitual sleep duration between 7 and 9 hours, a body mass index (BMi) between 18 and 29.9, normal sleep quality (Pittsburgh Sleep Quality Index score <5), no daytime sleepiness (Epworth Sleepiness Scale score <10), no depression symptoms (Beck Depression Inventory II score < 10), no medications (except for hormonal contraceptives), and no sleep disorders. For at least a week prior, participants followed a regular 8-hour sleep schedule of their own choosing at home, wore a wrist activity monitor, and maintained a daily sleep diary. Participants had a blood sampling catheter inserted on the afternoon of admission. After a 9-hour sleep episode centered at the participant’s sleep average times from the week prior, the participant was awakened for a 15-hour Baseline Day in the laboratory. They were served three regular meals, were allowed to shower in the morning, and spent the day in their study room pursuing sedentary activities such as reading, listening to music, etc. Hourly blood sampling began just after wake time. At the end of the Baseline Day a second 9-hour scheduled sleep episode occurred during which hourly sampling continued. Upon waking the next morning, participants remained in bed for a 39-hour constant routine (CR, see below for description). At the end of the CR the participant had a 12-hour recovery sleep episode during which hourly blood sampling continued. The study ended the following early afternoon. See Figure 3A for a schematic of the study protocol.

[0149] In the second study, nine adults (6 men, 3 women) between ages 18 and 25 were recruited for a 4-day study of sleep and brain structure. Inclusion / exclusion criteria were the same as in Study 1 , with the exception that they had to have a usual bedtime between 22:00 and 24:00 and a usual wake time between 06:00 and 09:00. After an 8- hour sleep episode, the participant was awakened and began a CR. A blood sampling intravenous catheter was inserted shortly after waketime, and hourly blood sampling began and continued for the next 36 hours. After 36 hours, the CR was ended, the catheter removed, and the participant escorted out of the CCI to the BWH MagneticResonance Imaging Research Center where the study continued. The first 24 hours of samples from the CR were used in the proteomics analysis (Figure 3A).

[0150] Overall, the average age of the 17 participants was 23.9 ± 3.9 years (mean ± standard deviation), their average BMI was 22.9 ± 2.0, and they were on average “neither” types based on their Morningness-eveningness scores (55.3 + 6.5). Sleep times during the week prior to study averaged from 22:31 ± 0:58 to 07:00 ± 0:52.Constant Routine

[0151] The CR consists of a regimen of enforced semi-recumbent wakefulness (sitting at a -45° angle in bed) in dim light, with nutritional intake divided into identical hourly snacks. Trained staff remain with the participant to ensure compliance and maintain wakefulness; waking electroencephalographic (EEG) data is collected to verify wakefulness. The CR ensures that specimens are collected under conditions of controlled posture, activity level, feeding, and sleep-wake state so that any observed rhythm icity in the data is due to endogenous sources rather than due to periodic behaviors (such as postural changes, sleep-wake state, or eating) or changes in the environment.Light Conditions

[0152] Lighting was provided by ceiling-mounted fluorescent lamps (T8 or T12 lamps with CCT of 4100K, Philips Lighting Eindhoven, The Netherlands) transmitted through ultraviolet (UV)-shielding ceiling filters (Lexan, GE Plastics, Pittsfield, MA). All lighting was always controlled by the experimenters and participants had no access to any other lighting.[01 S3] For Study 1 , lighting on the admission day was ~0.23 W / m2('-89 lux) at 137 cm from the floor facing the walls and had a maximum of 0.48 W / m2(150 lx) at 187 cm from the floor facing the ceiling anywhere in the room. Throughout Baseline Day and CR, lighting was set to -0.0087 W / m2(-3.3 lux) at 137 cm from the floor facing the walls and had maximum of 0.048 W / m2(15 lux) at 187 cm from the floor facing the ceiling anywhere in the room. Throughout each of the scheduled sleep episodes, all lighting was turned off (complete darkness).

[0154] For Study 2, Sighting at admission was -0.23 W / m2(-89 iux) at 137 cm from the floor facing the walls and had a maximum of 0.48 W / m2(150 lx) at 187 cm from the floor facing the ceiling anywhere in the room. Beginning ~6h before bedtime on the admission day, ambient lighting was reduced to -0.0087 W / m2(-3.3 lux) at 137 cm from the floor facing the walls and had maximum of 0.048 W / m2(15 lux) at 187 cm from the floor facing the ceiling anywhere in the room. The same light level was used throughout the CR. Throughout the 8-hour scheduled sleep episode, all lighting was turned off.Blood Sampling

[0155] A CCI research nurse inserted a 20-gauge intravenous catheter into a forearm vein for blood collection from each participant. The catheter was connected to a triplestopcock manifold (Cobe Laboratories Inc., Lakewood, CO) via an intravenous loop with a 12-foot small-lumen extension cable (Liberty Medical, Inc.) so that blood could be sampled from outside the room while the participant was sleeping. Between samples, a solution of 0.45% saline with 5,000 lU / liter of heparin was infused at a rate of 40 mL / hour to maintain patency.

[0156] Blood was collected hourly, and a 2 mL aliquot placed into a 3 mL EDTA vacutainer tube. The tube was inverted 8-10 times to ensure the EDTA was mixed with the blood, and then the tube was centrifuged at 3,000 rpm for 10 minutes at room temperature. A 200 pL aliquot of plasma was placed into a 2 mL microtube and frozen at -80°C. Every other sample was subjected to proteomics analysis in the present study to reduce overall costs.

[0157] An additional 1 mL of blood from each sample was put into a separate 3 mL EDTA vacutainer tube. Plasma from that tube was pipetted into a different microtube and frozen at -20°C for melatonin assay. Melatonin assays were performed by Solidphase, Inc. (Portland, ME) using the Buhlmann radioimmunoassay (NovoLytiX GmbH, Witterswil, Switzerland). This assay, based on the Kennaway G280 anti-melatonin antibody, has a sensitivity of 0.84 pg / ml, a range of 1-81 pg / ml, an intra-assay precision of 6.7%, and an inter-assay precision of 10.4%.Ethical Approval

[0158] Protocols were reviewed and approved by the Partners Health Care (now Mass General Brigham) Human Subjects Committee and conducted in accordance with the principles outlined in the Declaration of Helsinki. Each participant gave written informed consent prior to study.Protein Quantification

[0159] Plasma was assayed using the SomaScan aptamer-based multiplexed platform (SomaLogic Inc., Boulder, CO), which utilizes aptamers and hybridization to quantify proteins from plasma (L. Gold, et al., PLoS One. 2010 Dec 7;5(12):e15004, the disclosure of which is hereby incorporated by refernce). The platform is designed so that protein levels can be measured over a large range of concentrations. It includes both extracellular and intracellular proteins with soluble domains of membrane proteins. SomaScan assays have shown validity and reproducibility as well as stability. SomaLogic also conducts data quality controls at the sample and protein level to adjust for variability between and within samples and provides population-based normalized outputs of relative protein expression levels. Detailed information on SomaLogic’s quality control (QC) technique can be found on the manufacturer’s website. Briefly, QC procedures use pooled matrix-matched samples (e.g., plasma, serum) run in the SomaScan Assay alongside clinical samples to quantify the quality of each assay run by determining the accuracy of the median replicate signal for each SOMAmer reagent compared to the reference. QC check is performed after hybridization normalization, intraplate median signaling normalization, plate scaling, calibration, and adaptive normalization to a reference have been applied. SomaLogic provided two output files, each with different levels of population-based normalization, whereby the present study utilized the most normalized output based on our sample distributions. This study used the SomaScan platform of -'7,000 proteins. After removal of non-human proteins (e.g., mouse), a total of 6,916 proteins were quantified and analyzed. We further excluded samples that were flagged by SomaLogic as potential outliers.Statistical Analysis

[0160] We utilized the concept of internal time by calculating DLMO for each participant in each study condition. This involved fitting a 3-harmonic curve to the melatonin data, using a periodicity ranging from 24 to 24.3 hours with a step size of 1 minute. Subsequently, we determined the time at which the rising melatonin level reached 25% of the level of the fundamental harmonic. This DLMO marker (DLMO 25%) served as an indicator of the participant’s internal circadian time in each condition (Figures 3B and 3C).

[0161] To identify rhythmic diurnal and circadian proteins, we used a differential rhythmicity analysis based on a cosinor model with mixed effects that accounted for baseline changes between conditions and between participants. We divided the data into four “conditions”. In Study 1 we had a 24-hour Baseline sleep-wake day; the first 24 hours of the CR; and the final 15 hours of the CR plus the first 9h of the recovery sleep. In Study 2 we used the first 24 hours of data from the CR (Figure 3C).

[0162] Next, we applied cosinor analysis to estimate acrophases (peak times) and amplitudes of the protein profiles. The equation for one protein with n-harmonic model is:where tDLMO+represents number of hours after DLMO, which we use as internal time. We tested the null hypothesis that the sine and cosine terms were equal to zero, allowing us to identify proteins exhibiting daily rhythmicity. Proteins that surpassed a false discovery rate (FDR) rhythm threshold (rhy_fdr) of 0.01 and met a minimum amplitude requirement (amp_cutoff) of 0.1 in at least one condition were classified as rhythmic (but not necessarily circadian).

[0163] Next, we compared variations across the four conditions for proteins identified as rhythmic. We tested the null hypothesis of equal sine and cosine terms for each protein across conditions. If the null hypothesis were rejected under an FDR threshold (comj'dr) of 0.05, it indicated a significant difference in rhythms between conditions for that protein, suggesting a behavioral or environmental source such as posture, sleepwake state, eating, or light-dark changes causing rhythmicity. Conversely, if the nullhypothesis could not be rejected, the protein was considered to have a consistent rhythm between conditions, suggesting an endogenous circadian origin. We adjusted the FDR threshoids (rhyj'dr and com_fdr) and changed the amplitude requirement (amp_cutoff) to strike for a balance between sensitivity and specificity, ensuring identification of circadian proteins while rejecting false positives (Figure 4).

[0164] To accommodate variations in baseline expression between conditions and participants, we employed linear mixed models. These models were used to estimate and account for individual-specific baseline levels of protein expression, capturing any inherent differences in baseline between conditions and between participants. This adjustment enabled us to discern genuine circadian rhythm icity by focusing on variations attributed to the endogenous circadian system rather than overall differences In protein level between individuals or conditions.

[0165] We performed cosinor analysis with both one and two harmonic components to accommodate for complex circadian behaviors that could manifest higher harmonics. This approach enabled us to capture additional variations in the protein expression patterns, thereby enhancing sensitivity.

[0166] We used a soft-dtw barycenter technique to compute the average circadian expression profiles (Figures 5A and 5B) (A. Sarda-Espinosa R package vignette. 2017: 12: 41 , the disclosure of which is hereby incorporated by reference). A regulation value of 10 was utilized, except for the final 15 hours of CR data and the initial 9 hours of recovery sleep data in Study 1. We used a regulation five times stronger for this last condition to compensate for fewer data points. To assess stability of the barycenter, we implemented a leave-one-participant group-out strategy, providing insights into robustness and reliability in capturing overall patterns across participants.

[0167] We performed pathway analysis on identified circadian proteins with peak expression occurring within ±1 hour intervals, utilizing the GO_Bioiogical_Process_2021 , Reactome__2022, and KEGG__2021__Human databases using the Gene Set Enrichment Analysis (GSEA) computational method. To map pathway activity over the 24 hours, we normalized combined scores along the time axis.Results[0i 68] Overall, we assayed 432 samples from the 17 subjects: Study 1 , 24-hour sleepwake day (122 samples from 8 participants); first 24 hours of CR (112 samples); the final 15 hours of CR and first 9h of recovery sleep (91 samples); Study 2, first 24 hours of data from the CR (107 protein samples from 9 participants).

[0169] Utilizing the default hyperparameters presented in Figure 4, our one harmonic analysis identified 804 proteins exhibiting rhythmic patterns. Among these, 293 displayed consistent 24-hour rhythms between conditions and participants, indicating a circadian origin. In our two harmonics model, we discerned 895 rhythmic proteins, with 328 showing circadian characteristics. Notably, 259 proteins identified using the two harmonics approach were not detected in the single harmonic method (Figure 5A). Combining insights from both models, among the 6,916 proteins we identified a total of 1 ,063 (15%) proteins with rhythmic tendencies. Among these, 431 (6.2%) exhibited clear circadian rhythms across varied conditions and participants. Further characterization of these circadian proteins by amplitude and phase revealed deeper insights into their oscillation strength and timing (Tables 1A and 1 B). Figure 5B displays a selection of the most robust circadian proteins, emphasizing their inherent oscillatory patterns along with their respective amplitudes and phases.

[0170] Among the most robust circadian proteins were pro-opiomelanocortin (POMC) and parathyroid hormone (PTH), two well established circadian hormones (Figure 5B). POMC is a precursor of adrenocorticotropic hormone (ACTH), and peaked 11.3 hours after DLMO, a time consistent with the well-known morning peak of cortisol. PTH peaked 2.1 hours after DLMO, although a second, smaller peak was present under entrained sleep-wake conditions as has been described for circulating levels of the hormone.

[0171] Previously unknown circadian proteins that were identified include glutathione S-transferase A2 (GSTA2); four-jointed box protein 1 , a protocadherin involved in cancer progression; chemokine (C-C motif) ligand 15; kallikrein-14, a protein involved in skin desquamation and prostate cancer; tissue-type plasminogen activator; espin, a microfilament binding protein; and GPR37.

[0172] When we examined the time at which the circadian proteins peaked, we found that their acrophases were not distributed randomly across the 24-hour day. Rather, more proteins peaked in the late afternoon / evening, during the wake maintenance zone whenthe circadian rhythm of temperature is at its peak and circadian drive for alertness is highest (Figure 6). A second, smaller peak of circadian protein acrophases was observed during the latter half of the habitual sleep episode, in the early morning when the endogenous circadian rhythm of temperature is at its nadir and the circadian rhythm of sleep propensity is at its peak (Figure 6).

[0173] Finally, we carried out pathway analysis for proteins peaking within ±1 hour, adding additional insight into many of the novel endogenous circadian proteins identified in our study. We next grouped the pathways into four segments by the time within the 24- hour day at which they peaked. This is presented visually in Figure 6. Pathways peaking in the evening are diverse and include immune (T cell development), prostaglandin metabolism, chondrocyte maturation, and protein localization to axons. Among the most significant pathways, positive regulation of nucleotide-binding oligomerization domain containing 2 (NOD2) signaling, an inflammatory pathway, and two heat acclimatation pathways peaked in the morning, a time when few pathways were active (Figure 6). Pathways active in the late night / early morning included innate antibacterial humoral responses, anion transport, negative regulation of metallopeptidase activity, mucosal immune responses, glial and endothelial cell regulation.Conclusions

[0174] We described the physiology of the blood circadian proteome in humans by measuring nearly 7,000 proteins in samples collected every two hours from 17 healthy participants using the gold-standard constant routine (CR) protocol. Applying both a single and dual harmonic method of circadian analysis to the data, we identified 1 ,063 diurnal proteins, 431 of which exhibit endogenous circadian rhythm icity. Thus, at least ~10-15% of circulating proteins are regulated by the circadian system, with proteins peaking at different times across the 24-hour cycle, although most commonly in the late afternoon or early evening hours as illustrated in Figure 4. Our approach to studying the blood circadian proteome in humans allowed us to identify and characterize proteins with circadian behaviors beyond simple sinusoidal rhythms. The distinction between diurnal and endogenous circadian proteins, coupled with assessment of amplitude and phase,provides insights into the complexity, stability, and temporal characteristics of the multi- oscillatory circadian timing system in humans.

[0175] Using the CR protocol, we were able to distinguish endogenous circadian proteins from those with diurnal fluctuations, many of which could be driven by the timing of sleep, activity, posture, or food intake. This distinction, often overlooked in studies of human circadian rhythmicity, is critical for understanding clock dysfunction. Rhythms driven by behaviors rather than by endogenous circadian clocks would be expected to be altered in shift workers, patients with circadian rhythm sleep-wake disorders, and others (such as patients in Intensive Care Units); understanding the distinction between endogenous circadian rhythms versus diurnal rhythms is critical to be able to use proteins to identify endogenous circadian phase as well as to identify the presence of circadian disruption. Reassuringly, POMC, the precursor of ACTH and a driver of cortisol, had one of the most robust patterns, peaking almost antiphase to DLIVIO, as expected.

[0176] Other circadian proteins identified included tissue-type Plasminogen Activator (tPA), which is involved in fibrinolysis and is associated with sleep apnea. Interestingly, GSTA2, an enzyme involved in hematopoiesis and detoxification, notably for chemotherapeutic agents such as paclitaxel and in the hepatotoxicity of selected medications, peaked late at night, of possible significance for chrono-pharmacology. Chemokine (C-C motif) ligand 15, an important chemotaxis agent for neutrophils, monocytes, and lymphocytes, also peaked late at night, and may, in addition to the modulatory effects of cortisol, explain well-known circadian variations in white blood cell subtypes across the 24-hour day. Carboxypeptidase B, a protease involved in the biosynthesis of neuropeptides and peptide hormones is primarily secreted by the pancreas, is involved in complement activation and other endocrine functions, and peaks in the afternoon. Prosaposin / prosaptide receptor GPR37, a brain receptor for saposins, important neurotrophic and glioprotective factors for Parkinson disease, and SPARC- related modular calcium-binding protein 1 (SMOC-1) a biomarker of tau involvement in Alzheimer’s disease, were also strongly circadian.

[0177] Our study of nearly 7,000 plasma proteins allowed us to probe the output of many organs and physiologic functions, showing the feasibility of large-scale proteomic studies of circadian rhythmicity. Grouping the proteins by pathways revealed additionalinsight. Pathways regulating T cell differentiation and activation in the thymus occurred in the late afternoon. Cell numbers in blood as well as T cell egress from the thymus are influenced by sleep and circadian timing, peaking in the evening and early night, a few hours after this pathway is activated. Prostaglandin metabolism and chondrocyte maturation, strongly circadian processes, also peaked at this time. In contrast, pathways relating to endothelial cell proliferation and repair peaked in the early morning, reflecting circadian and organ specific modulation. Understanding the timing of such pathways may provide new insights into the best time of day to administer certain medications or treatments. A notable finding was the strong endogenous circadian regulation of the heat acclimatation pathway, which peaked in the late morning when most other pathways are silent. This may reflect anticipation ofthe warmest time of day, a timely finding considering the threat of global warming. Similarly, the N0D2 pathway, important for autoinflammatory syndromes such as Chron’s disease, Blau syndrome, and N0D2- associated autoinflammatory diseases, is also active in the late morning.

[0178] Proteomics is increasing as a field of inquiry, yet studies rarely if ever control for time of day or circadian effects, effects that the present findings demonstrate could be confounding, or if considered could increase power for discovering new associations. Only one study has explored circadian effects using a smaller panel of 1 ,300 proteins in a simulated shiftwork protocol. It is becoming more evident that circadian abnormalities are present (and even at the core of) in many disorders from psychiatric to neurodegenerative, thus understanding how rhythmic proteins may be altered in these conditions has its own value. Two proteomics platforms are available and offer comparable performance, Olink, an antibody-based platform, and Somalogic, an aptamer-based technology. However, correlations with mass spectrometry or enzyme- linked immunosorbent assay (ELISA) vary, so that any individual result must be interpreted cautiously. Nonetheless, integration of these results with genomic data has generally validated many targets by revealing cis-pQTLs close to each gene of interest, and these findings are now being integrated with genomic data through Mendel ian randomization studies. Similar experiments may, in the future, allow convergence of natural variation associated with morningness / eveningness or other diseases with specific pathways or organ specific clocks.

[0179] The discovery of proteins peaking at various times aiso yields the ability to buildi an algorithm to predict central circadian clock phase with a single blood sample, a circadian “time stamp”. Further, in some cases validation has been done in entrained conditions that cannot distinguish diurnal from circadian and thus may not accurately reflect the status of the circadian system. One exception has used gene expression within a single cell population, monocytes, and has shown good correlation with DLMO. It is likely that noninvasive measures of physiology through wearables, together with cell specific measures of circadian rhythm icity and global variation of the human proteome and metabolome will lay the foundation for improved circadian time stamp methods.

[0180] A major strength of our study was the use of the CR protocol which allowed us to distinguish between diurnal rhythms (likely driven by rhythmic behaviors) and endogenous circadian rhythms. Our approach revealed hundreds of plasma proteins that show endogenous circadian rhythm icity, from a wide range of organ systems and physiologic functions.COMPUTATIONAL MODEL SURROGATE OF DIM LIGHT MELATONIN ONSET

[0181] All fundamental aspects of human physiology, metabolism, and behavior display 24-hour rhythms. Although synchronized by a master clock located in suprachiasmatic nuclei of the hypothalamus, almost every cell of the body contains its own circadian oscillator that contributes to the daily rhythmicity of a large variety of physiological and metabolic activities, including immune responses and drug detoxification as well as renal, hepatic, pancreatic, endocrine, reproductive, respiratory, and cardiovascular functions. In fact, at least 40% of protein coding genes show daily rhythms in expression in a tissue-specific manner in humans.

[0182] Given this ubiquity, it is not surprising that measuring circadian phase conveniently in humans will have many clinical applications across multiple clinical areas, a subfield called chrono-medicine. For example, metabolism (pharmacokinetics) and effects (pharmacodynamics) of many, especially short-half-life, drugs change over the 24-hour day. Similarly, outcomes in areas as varied as survival after open-heart surgery, efficacy and tolerance of chemotherapy, wound healing, antibody response to vaccination, and effectiveness of statins or acetylsalicylic acid application forcardiovascular diseases all vary with time of day. It is also notable that abnormal circadian rhythms are implicated in several psychiatric disorders including schizophrenia, depression, and bipolar disorder.

[0183] In addition to the applications in general medicine, there is an urgent need for a practical way to assess circadian timing accurately in sleep medicine, a discipline that requires diagnosis of millions of patients with chronic insomnia (10% of the population) or hypersomnia (4% of the population), for which current standard treatments are behavioral therapy, hypnotic compounds, or stimulants. In fact, a recent study has found that up to one quarter of insomnia patients have an abnormal circadian phase, 14 which leads to entirely different therapies (for example melatonin or light administration), properly timed with the circadian phase information. The circadian rhythm sleep disorders Advanced Sleep-Wake Phase Disorder (ASWPD) and Delayed Sleep-Wake Phase Disorder (DSWPD) are so-named because it is hypothesized that the abnormal sleep timing is due to an abnormal circadian rhythm timing. However, recent studies have found that only half of adolescent or adult patients who meet the diagnostic criteria for DSWPD actually have a delayed phase. Similarly, recent studies have found that abnormal Multiple Sleep Latency Tests (MSLT), in which Sleep Onset REM Periods (SOREMPs) are observed, the hallmark for the diagnosis of narcolepsy, can be confounded by abnormal circadian phase timing — consistent with earlier observations in healthy participants18. As an example, the probability of a false positive (for SOREMPs) MSLT is increased 8~fold in shift workers, and more than half of positive MSLTs reported in the context of a narcolepsy type 2 diagnosis are not stable traits but instead suggest a circadian abnormality. In both cases, the patient would be provided (ineffective) pharmaceutical treatment for narcolepsy, when instead they had a circadian rhythm sleep-wake disorder that could be treated with melatonin, a melatonin agonist, or light therapy. Thus, there is also an urgent need for accurate circadian timing biomarkers for the correct diagnosis and treatment of many sleep disorders patients. In addition to usefulness in the diagnosis of sleep disorders, circadian timing biomarkers that are based on multiple rhythmic features will allow' for detection of internal desynchrony, e.g., when rhythms in genes and / or proteins are out of sync with each other and / or with the timing of the central clock. There is evidence that such internal desynchrony occurs in shift workers.Results

[0184] Circulating protein measurements were obtained using SomaScan (Slow Off- rate Modified Aptamer-Scan), as described herein. Briefly, we collected samples every 2 hours and assess each sample for 5,500 plasma proteins using the Somascan technology platform. The Somascan method uses artificially developed oligonucleotides that interact specifically with selected proteins. What is remarkable about this technology is that it can be multiplexed so 5,500 different proteins (current panel size) can be measured simultaneously in a small volume.34 It also has the ability to measure a large, dynamic range of concentrations (from fM to pM) simultaneously without having to use separate dilutions for each assay, allowing for multiplexing (Figure 7).

[0185] We developed a panel of rhythmic proteins that comprise a circadian phase biomarkers, and validated the protein panel against the gold standard dim light melatonin onset (DLMO) measure of circadian phase. We will build upon our data by studying each participant under the rigorous, controlled conditions of the Constant Routine, under their regular sleepwake conditions, and under irregular sleep-wake conditions (simulating changes in sleep duration and timing between weekdays and weekends).

[0186] Utilizing the Somascan panel, we performed a rhythm icity analysis to determine the number of proteins that have a circadian waveform. Among the 1300 proteins, we identified 129 proteins that showed circadian rhythmicity (FDR p value < 0.05). Having confirmed the circadian rhythimicity, we trained a circadian time predictor on the entirety of the intersection of proteins (n --■1013) from our 1300 Somascan protein panel.

[0187] To make our model performance robust to conditions such as inverted sleep (daytime sleep and night time awake), we also appended shared protein (n=1013) data to our training cohort derived from 70% of Somascan data, which contains both constant routine and inverted sleep conditions, setting aside 30% for a validation cohort. In our cohorts, the morning samples were drawn after habitual wake time and reflected the internal circadian time, as validated by peaks in ACTH (adrenocorticotrophic hormone) and proopiomelanocortin. The model was then evaluated for performance in two independent validation cohorts with cohort 1 also containing inverted sleep timepointsand cohort 2 with constant routine (Figures 8A & 8B). The median absoiute error (MdAE) was 1 .24 and 1 .69 hours across cohort 1 and cohort 2 respectively.

[0188] We also had access to a 5500 protein panel which we tested out. We used this panel on plasma derived from 9 individuals who were sampled every 2 hours over a 36- hour Constant Routine protocol. While 70% of the data was used as a training and analysis set for building a circadian time predictor, the rest was retained as an untouched validation set. At first, as detailed above, we verified circadian rhythmicity of the 5500 somascan protein panel utilizing harmonic regression. This analysis revealed 379 plasma proteins (FDR p=0.05) to be circadian regulated: strikingly, ACTH and proopiomelanocortin peaked in the early morning and tapered as the day progressed, consistent with established circadian time stamps (Figure 9). Other proteins of interest with diurnal rhythmicity included thyroid / parathyroid hormone, PGD2 synthetase ghrelin and pancreatic hormone precursor.

[0189] While the harmonic analyses in this pilot study revealed previously unknown circadian rhythmicity of certain proteins, we also trained a circadian time predictor PlasmaTime using the 5500 Somascan panel. In the process, a multi-protein machine learning regressor was trained on the Somascan 5500 plasma protein panel utilizing a lasso shrinkage model with L1 regularization. The 5500 protein panel performed robustly with a high circular Pearsons correlation at 0.94 between actual and predicted times (Figure 10), while also predicting the circadian time to within an hour. The MdAE in a validation set consisting of 53 timepoints was 0.80 hrs (0:48 minutes clock time). In 80% of the samples prediction error was less than 1 :25 hrs and in 98% of the samples, the prediction error was less than 2hrs (Figure 10). The most common proteins picked by the lasso model (present in at least 70% of 100 instances of training on the 5500 somascan protein panel) are listed in the table of Figure 11 in order of their frequencies.

[0190] As noted above, the rationale for our project goal of developing a single timepoint circadian phase biomarker is the need for such a clinical tool and the lack of one currently. For research applications, blood or saliva samples are collected hourly, ideally for a full 24 hours. Because melatonin secretion is suppressed by light (even light of indoor intensity), the subject must remain in dim light (ideally <20 lux) conditions for the entire time. A shortened approach to assessing circadian phase is to collect samples(in dim light) for several hours around the time when melatonin secretion is predicted to begin so that the onset of secretion can be determined (the dim light melatonin onset, DLMO), although this still requires 5 or more hours of sampling in dim light. In order to test whether such sample collection could be done by patients in their own home, we carried out a study to compare the DLMO estimates from hourly saliva samples collected by participants in their home to a DLMO estimate collected in our highly controlled research laboratory on the following evening. While participants in our original study did not have to meet criteria for a circadian rhythm sleep wake disorder, we selected individuals with a complaint of either difficulty falling asleep at their desired time, or difficulty awakening earlier than desired. Using a fixed threshold phase marker, we found a strong correlation (r = 0.85, n ~ 18, p < 0.0001 ) between the DLMO estimate collected by the participant at home and that collected in our laboratory on the following evening (Figure 12). This established a method and the feasibility of assessing DLMO in patient’s homes.[0i 91] We have introduced this method in the Sleep Center at Brigham and Women’s Faulkner Hospital, where patients with circadian rhythm sleep wake disorders can selfpay for this assessment. Since its introduction, all the DLMO profiles from patients who have collected saliva samples at home have been of high quality. In the example presented in Figure 13, the patient’s DLMO is more than 2 standard deviations later than that seen in healthy adults, yet it was more than 5 hours earlier than their usual bedtime (of -06:00),

[0192] Experimental Participants. Healthy men and women, 21-35 years of age, free from any acute, chronic, or debilitating medical conditions, were recruited from the greater Boston community and selected for Study 1. Screening (detailed in Human Subjects) includes rigorous medical, psychological, and sleep-wake evaluations. An important aspect of our participant selection criteria includes ensuring ail participants have maintained a very regular sleep-wake / dark-light schedule with sufficient time for sleep for at least 2 weeks immediately before study to ensure they are stably entrained and well- rested.[Oi 93] A total of 200 adults who meet ICSD-3 criteria for Advanced Sleep-Wake Phase Disorder or Delayed Sleep Wake Phase Disorder are for the study, 100 each at Brigham and Women’s Hospital and Stanford University.

[0194] Melatonin data. The recording of the circadian rhythm of melatonin is essential, as it is considered a “gold standard" way to assess central circadian phase. The melatonin rhythm has been demonstrated to be the most precise and accurate marker of the human circadian pacemaker 60-62. In Study 1 , samples of blood were collected every hour through the blood collection system described above. Blood samples (approximately 1 mL) were transferred to small (3 cc) vacutainer tubes and immediately centrifuged at 4°C; the resulting plasma will be pipetted into polystyrene tubes and frozen at -80°C until analysis. In Study 2, saliva samples (approximately 1-2mL) were collected once per hour into polystyrene tubes in the patients’ homes and frozen until transfer Melatonin assays will be performed by Solidphase (Portland, ME) using the Buhlmann radioimmunoassay (ALPCO Diagnostics, Salem NH). This assay is based on the Kennaway G280 antimelatonin antibody (210), has a sensitivity of 0.84 pg / ml, a range of 1-81 pg / ml, an intraassay precision of 6.7%, and an inter-assay precision of 10.4%. We have used Solidphase to assay melatonin samples for the past decade and have found them to be efficient and reliable.

[0195] SOMALogic proteomics. In Study 1 , samples of blood will be collected every two hours through the blood collection system described above. Blood samples (approximately 1 ml.) will be transferred to small (3 mL) EDTA vacutainer tubes and mmediately centrifuged at 4°C; samples (150pl) of the resulting plasma will be pipetted into polystyrene tubes and frozen at -80°C until analysis. In Study 3, stored samples that were originally collected in the same way will be selected for inclusion. In Study 2, a single blood sample will be collected from each patient-participant directly into an EDTA vacutainer tube and processed as above. These plasma samples will be sent to Somalogic to assay 5,500 proteins.00Table IB. Statistics of Proteins Identified as Rhythmic.C0L1A1 0.001924414 0.006984202 0.552804317 0.463644461 0.048742003 0.049244383 0.015901928 4.226414881 4.137795145 10.57999987 ANGPTL1 6.56E-42 7.56E-40 0.00731308 0.010021876 0.090003325 0.091891393 0.009006779 19.85606561 19.85387847 8.030997907 TREML1 0.000782488 0.000378519 0.013937693 0.004208745 0.033671544 0.034143293 0.01220599 21.34729965 21.53464663 12.72503326 KLK13 0000781456 0.005951286 0.054854707 0.084951003 0.035865848 0.035939851 0.002570963 4.176008363 4.112129081 16.63784272 SVEP1 2.65E-09 3.93E-08 0.002801918 0.004284477 0.02483709 0.025643636 0.002428845 15.55856877 15.66029031 9.304978703 COL6A3 3.02E-17 8.81E-15 3.19E-10 9.87E-09 0.038103671 0.038688605 0.007054704 5.544442897 5.512956475 12.81591115DNAJB9 1.08E-10 8.58E-10 6.64E-09 4.64E-08 0.042431357 0.045414892 0.021941925 17.88329032 17.78485007 I.414780934 TPMT 5.69E-08 2.27E-07 0.000216959 0.00215436 0.026934057 0.02853838 0.015286237 6.585909373 6.519968225 15.99793967 FGFBP.3 0.000779799 0.000556819 0.00615443 0.01149974 0.025199944 0.025709804 0.018160376 4.515828744 4.498716402 II.9075122.3 ERN1 1.46E-07 4.45E-07 1.17E-05 7.73E-06 0.021306882 0.020948183 0.004141346 18.2814743 18.39518403 5.712966981 ASL 0.001718001 0.000352038 0.006002401 0.001484367 0.050148469 0.051018609 0.030409287 9.864165676 9.849556547 21.0987275 NAGS 0.001294009 0.039333493 0.00408212 0.046523401 0.01625377 0.016204154 0.004528941 21.33109201 21,53250606 12.637472.7 ATF6 7.05E-06 0.000971.387 0.175910544 0.536233848 0.032642916 0.032561899 0.001280312 18.76216756 18.72798716 .3.253443473 CREB3L4 6.01E 20 4.44E-19 6.53E-14 2.67E-13 0.031342938 0.032095806 0.011595624 8.064698817 7.969986776 12.43261073ARHGAP25 0.001252905 0.006326654 0.000655187 0.003044791 0.028115 0.028509479 0.023308294 21.42454896 21.37660687 I.684025003 NHEJ1 0.002581233 0.025481421 0.00486856 0.022002357 0.027411393 0.028640447 0.010726484 20.10536131 20.1105492 4.497557311 GRB10 0.003530093 0.027187783 0.0061991 0.026054269 0.080423701 0.08253902 0.03656503 20.28767039 20.30405873 2.688426198 ADH7 3.93E-06 0.000503754 0.031785182 0.15205957 0.028270856 0.028583736 0.00189223 15.6915387 15.71492814 0.133739045 ATF6B 1.24E-06 1.23E-07 0.002022385 0.000540157 0.023777713 0.023945489 0.012106537 18.92774873 18.96292025 8.593227454 WFDC2 1.44E-15 1.70E-15 2.29E-07 5.00E-07 0.056905252 0.057876941 0.020918748 5.668550848 5.66199364 12.17438451V3INA 2.19E-08 1.22E-07 9.37E-05 0.000189112 0.030608511 0.032466197 0.011723656 8.970410155 9.13229876 18.72277938 RBFOX2 2.75E-05 7.15E-06 0.005562269 0.002229374 0.022618287 0.023615495 0.012212965 7.507355853 7.557154008 15.50370292 RBM19 0.070762642 0.006313544 0.059739205 0.008076743 0.010995816 0.012240976 0.010255119 12.40914245 12.50757465 12.30944809 CD59 1.64E-15 1.28E-13 2.94E-08 6.29E-07 0.053606521 0.054408702 0.014295565 4.986330964 4.893008746 10.80035973 FABP1 4.75E-22 3.30E-30 5.26E-07 1.81E-13 0.115308173 0.120881895 0.043994508 5.647465512 5.440101781 II.44503666 ARFGAP1 0.000309874 0.008281729 0.000196612 0.004128639 0.011463581 0.011053226 0.009379056 8.522767693 9.008459114 0.59108063RBM23 0.01269737 0.000877803 0.285151589 0.022416752 0.028812733 0.0292.83495 0.010410569 7.867054642 7.915749343 10.91890137 SMAP1 0.008387979 0.032645752 0.018563221 0.047752259 0.061647081 0.063716153 0.031821548 20.88783636 20.93165461 1.751543654CEINT1 0.002581233 0.020055827 0.002617456 0.010401015 0.058180266 0.060737434 0.02626595 21.54994197 21.65937235 4.530924618 TNS2 2.41E-05 6.72E-05 0.004130757 0.005067256 0.019600994 0.020568916 0.009139141 8.296053826 8.439384924 17.06570891 CRABP2 0.000668788 2.26E-05 0.805179058 0.08197254 0.042758626 0.039138995 0.017295183 6.648511155 6.558190711 6.69456117 LCN1 6.43E-06 0.000463689 0.071872831 0.323073489 0.033828884 0.034076231 0.009691534 6.83518982 6.758275323 10.80820726 DLX4 0.007897046 0.073646437 0.006446209 0.036019818 0.039852707 0.041722047 0.02023829'1 19.66591103 19.88018895 5.3 / 0887849 ARHGAP1 0009475593 0.007181388 0.010587805 0.004126743 0.016440664 0.018885452 0.009954393 5.305814607 4.781906421 9.572656595CBL 0.001908144 0.006056341 0.004807172 0.013215094 0.065040527 0.066825752 0.038392107 20.05328207 20.0411858 1.8382516 ETS2 0.000195945 0.002277033 0.103953657 0.16544715 0.021754272 0.023068376 0.004363052 16.64247941 16.52774176 1.949544233 MRRF 0.004099581 0.00928682 0.111145602 0.130246682 0.022623162 0.022318368 0.008490109 18.93305175 18.9086394 23.95432832 RUVBL1 1.02E-06 3.53E-07 7.47E-06 7.23E-07 0.019236882 0.019878825 0.00400311 18.18253857 17.83969022 20.54851508 A. PDF 5.44E-16 1.44E-13 0.007386045 0.018964691 0.075614143 0.075716773 0.00414619 7.777759986 7.881313494 23.53002383 TRA2B 6.18E-22 5.65E-23 2.71E-06 3.81E-08 0.103508334 0.103185704 0.017060232 7.330552221 7.317933178 9.808309441co cot'oMcb V373

Claims

WHAT IS CLAIMED IS:1 . A method of predicting circadian rhythmic phase of an individual, comprising: receiving, using a computational processing system, proteinaceous species measurements from a biological sample collected from an individual: and classifying, using the computational processing system, the biological sample as being associated with a rhythmic phase by entering the proteinaceous species measurements into a trained rhythmic phase classification model.

2. The method of claim 1 further comprising: determining a timing of a circadian rhythmic event of the individual based on the time the biological sample was collected and the rhythmic phase of the biological sample as classified by the trained rhythmic phase classification model.

3. The method of claim 2, wherein the circadian rhythmic event is determined to be delayed or advanced, the method further comprising: administering one or more modulators of circadian rhythm phase.

4. The method of claim 3, wherein the one or more modulators modulator or circadian phase comprises one or more melatonin receptor agonists or light therapy.

5. The method of any one of claims 2-4, wherein the circadian rhythmic event is dim light melatonin onset.

6. The method of any one of claims 1 -5 further comprising: correcting for individualism using the computational processing system, wherein correcting for individualism comprises: determining a proteinaceous species signature of the individual; determining, using the proteinaceous species signature, circadian- rhythmic-related species measurements deviated from a population norm: andremoving the deviated circadian-rhythmic-related species prior to entering the proteinaceous species measurements into the trained rhythmic phase classification model.

7. The method of any one of claims 1 -6 further comprising: correcting for individualism using the computational processing system, wherein correcting for individualism comprises: determining a proteinaceous species signature of the individual: determining an individual’s personalized center of the proteinaceous species signature; and adjusting the classification result using the individual’s personalized center based on a difference between a center of a population norm and the individual's personalized center.

8. The method of any one of claim 1 -7 further comprising: measuring proteinaceous species within a proteinaceous species sample, wherein the proteinaceous species sample is or is derived from the biological sample.

9. The method of claim 8, wherein measuring proteinaceous species comprises one or more of the following: ELISA, antibody arrays, DNA bar-coded antibodies, aptamers, chromatography, mass spectrometry, NMR, or electrophoresis.

10. The method of claim 8, wherein measuring proteinaceous species comprises: contacting the proteinaceous species sample with a panel of aptamers that are configured to bind select proteinaceous species; and sequencing the aptamers that bind a proteinaceous species.11 . The method of claim 8, wherein measuring proteinaceous species comprises: contacting the proteinaceous species sample with a panel of DNA bar-coded antibodies that are configured to bind select proteinaceous species; and sequencing DNA bar codes of the antibodies that bind a proteinaceous species.

12. The method of claim 10 or 11 , wherein the aptamers or the DNA bar-coded antibodies are configured to bind two or more of the following proteinaceous species: angiopoietin-related protein 1 , cathepsin F, pro-opiomelanocortin, prolactin, metalloproteinase inhibitor 4, hyaluronidase-1 , thyroid stimulating hormone, kallikrein-7, secretogranin-1 , cysteine-rich secretory protein LCCL domain-containing 2, hemopexin, zymogen granule protein 16 homolog B, pancreatic hormone, follistatin-related protein 3, neutrophil cytosol factor 1 , heparan-sulfate 6-O-sulfotransferase 2, cGMP-dependent 3’,5'-cyclic phosphodiesterase, HLA class I alpha chain G, lactotransferrin, proprotein convertase subtilisin / kexin type 9, secreted and transmembrane protein 1 , desmocollin- 2, succinate dehydrogenase assembly factor 2 -mitochondrial, and lysosomal protective protein.

13. The method of any one of claims 1 -12 wherein the trained rhythmic phase classification model comprises the use of one or more of: linear regression, polynomial regression, Cox proportional hazards regression, multiple linear regression, ridge regression, logistic regression, Lasso regression, stepwise regression, principal component analysis, Bayesian inference, elastic net, and random forest regression.

14. The method of any one of claims 1 -13 further comprising: training a classifier to yield the trained rhythmic phase classification model, comprising: collecting a plurality of biological samples for each individual of a cohort of individuals, wherein the plurality of biological samples span across multiple circadian phases; determining a circadian rhythmic event for each individual of the cohort; assigning each biological sample a time stamp relative to the circadian rhythmic event; measuring proteinaceous species within each biological sample to yield proteinaceous species measurements; andentering the proteinaceous species measurements and associated circadian rhythmic event time stamp into a classifier model to train the model to learn the relation between the proteinaceous species measurements and the circadian rhythmic event time stamp.

15. The method of claim 14, wherein the circadian rhythmic event is dim light melatonin onset.

16. The method of any one of claims 1 -15, wherein the individual has a medical condition selected from one of: one of blindness, narcolepsy, chronic insomnia, hypersomnia, advanced sleep-wake phase disorder, delayed sleep-wake phase disorder, irregular sleep-wake rhythm disorder, jet lag disorder, non-24-hour sleep-wake rhythm disorder, or shift work disorder.

17. A method of improving administration of a treatment affected by circadian rhythm phase, comprising: receiving, using a computational processing system, measurements of proteinaceous species from a biological sample collected from an individual, wherein the individual has diagnosis for a medical condition to be with treated with a medication that is affected by circadian rhythm; classifying, using the computational processing system, the biological sample as being associated with a rhythmic phase by entering the proteinaceous species measurements into a trained rhythmic phase classification model; determining a timing of a circadian rhythmic event of the individual based on the time the biological sample was collected and the rhythmic phase of the biological sample as classified by the trained rhythmic phase classification model; and administering the medication to the individual to treat the medical condition at an efficacious time based on the individual’s determined timing of the circadian rhythmic event.

18. The method of claim 17 further comprising: administering one or more modulators of circadian rhythm phase to modulate the timing of the circadian rhythmic event.

19. The method of claim 18, wherein the one or more modulators modulator or circadian phase comprises one or more melatonin receptor agonists or light therapy.

20. The method of any one of claims 17-19, wherein the circadian rhythmic event is dim light melatonin onset.21 . The method of any one of claims 17-20 further comprising: correcting for individualism using the computational processing system, wherein correcting for individualism comprises: determining a proteinaceous species signature of the individual; determining, using the proteinaceous species signature, circadian- rhythmic-related species measurements deviated from a population norm; and removing the deviated circadian-rhythmic-related species prior to entering the proteinaceous species measurements into the trained rhythmic phase classification model.

22. The method of any one of claims 17-21 further comprising: correcting for individualism using the computational processing system, wherein correcting for individualism comprises: determining a proteinaceous species signature of the individual; determining an individual’s personalized center of the proteinaceous species signature; and adjusting the classification result using the individuals personalized center based on a difference between a center of a population norm and the individual’s personalized center.

23. The method of any one of claim 17-22 further comprising: measuring proteinaceous species within a proteinaceous species sample, wherein the proteinaceous species sample is or is derived from the biological sample.

24. The method of claim 23, wherein measuring proteinaceous species comprises one or more of the following: ELISA, antibody arrays, DNA bar-coded antibodies, aptamers, chromatography, mass spectrometry, NMR, or electrophoresis.

25. The method of claim 23, wherein measuring proteinaceous species comprises: contacting the proteinaceous species sample with a panel of aptamers that are configured to bind select proteinaceous species; and sequencing the aptamers that bind a proteinaceous species.

26. The method of claim 23, wherein measuring proteinaceous species comprises: contacting the proteinaceous species sample with a panel of DNA bar-coded antibodies that are configured to bind select proteinaceous species; and sequencing DNA bar codes of the antibodies that bind a proteinaceous species.

27. The method of claim 25 or 26, wherein the aptamers or the DNA bar-coded antibodies are configured to bind two or more of the following proteinaceous species: angiopoietin-related protein 1 , cathepsin F, pro-opiomelanocortin, prolactin, metalloproteinase inhibitor 4, hyaluronidase-1 , thyroid stimulating hormone, kallikrein-7, secretogranin-1 , cysteine-rich secretory protein LCCL domain-containing 2, hemopexin, zymogen granule protein 16 homolog B, pancreatic hormone, follistatin-related protein 3, neutrophil cytosol factor 1 , heparan-sulfate 6-O-sulfotransferase 2, cGMP-dependent 3',5'-cyclic phosphodiesterase, HLA class I alpha chain G, lactotransferrin, proprotein convertase subtilisin / kexin type 9, secreted and transmembrane protein 1 , desmocollin- 2, succinate dehydrogenase assembly factor 2 -mitochondrial, and lysosomal protective protein.

28. The method of any one of claims 17-27 wherein the trained rhythmic phase classification model comprises the use of one or more of: linear regression, polynomial regression, Cox proportional hazards regression, multiple linear regression, ridge regression, logistic regression, Lasso regression, stepwise regression, principal component analysis, Bayesian inference, elastic net, and random forest regression.

29. The method of any one of claims 17-28, wherein the medical condition comprises cancer and the medication comprises an immune checkpoint inhibitor.

30. The method of any one of claims 17-28, wherein the medical condition comprises cancer and the medication comprises a chemotherapeutic.31 . The method of any one of claims 17-28, wherein the medical condition comprises high cholesterol and the medication comprises a statin.

32. The method of any one of claims 17-28, wherein the medical condition comprises hypertension and the medication comprises an alpha-blocker.

33. The method of any one of claims 17-28, wherein the medical condition comprises hypertension and the medication comprises a beta-blocker.

34. The method of any one of claims 17-28, wherein the medical condition comprises hypertension and the medication comprises an angiotensin II receptor blocker.

35. The method of any one of claims 17-28, wherein the medical condition comprises allergic dermatitis or rhinitis and the medication comprises an antihistamine.

36. The method of any one of claims 17-28, wherein the medical condition comprises arthritis and the medication comprises an NSAID.

37. The method of any one of claims 17-28, wherein the medical condition comprises arthritis and the medication comprises an antimetabolite.

38. The method of any one of claims 17-28, wherein the medical condition comprises asthma and the medication comprises a phosphodiesterase inhibitor.

39. The method of any one of claims 17-28, wherein the medical condition comprises clotting and the medication comprises a p2y12 inhibitor.

40. The method of any one of claims 17-28, wherein the medical condition comprises anticoagulation and the medication comprises a Factor Xa inhibitor.41 . The method of any one of claims 17-28, wherein the medical condition comprises gastroesophageal reflux disease and the medication comprises a proton pump inhibitor.

42. A method of identifying a subtype of delayed sleep-wake phase disorder, comprising: receiving, using a computational processing system, measurements of proteinaceous species from a biological sample collected from an individual, wherein the individual has diagnosis for delayed sleep-wake phase disorder; classifying, using the computational processing system, the biological sample as being associated with a rhythmic phase by entering the proteinaceous species measurements into a trained rhythmic phase classification model; determining a timing of a dim light melatonin onset of the individual based on the time the biological sample was collected and the rhythmic phase of the biological sample as classified by the trained rhythmic phase classification model; and determining that the individual is a circadian delayed subtype or a non-circadian subtype based on the timing of a dim light melatonin onset.

43. The method of claim 42 further comprising: determining that the individual is a circadian delayed subtype; and administering one or more modulators of circadian rhythm phase to modulate the timing of the dim light melatonin onset.

44. The method of claim 43, wherein the one or more modulators modulator or circadian phase comprises one or more melatonin receptor agonists or light therapy.

45. The method of claim 42 further comprising: determining that the individual is a non-circadian subtype, wherein the individual is not administered one or more modulators of circadian rhythm phase to modulate the timing of the dim light melatonin onset.

46. The method of claim 45 further comprising: administering a hypnotic agent to the individual.

47. The method of claim 46, wherein the hypnotic agent comprises one or more of: a benzodiazepine, a non-benzodiazepine sedative-hypnotic, an orexin receptor antagonist, a sedative antidepressant, or an antihistamine.

48. The method of any one of claims 42 -47 further comprising: correcting for individualism using the computational processing system, wherein correcting for individualism comprises: determining a proteinaceous species signature of the individual; determining, using the proteinaceous species signature, circadian- rhythmic-related species measurements deviated from a population norm; and removing the deviated circadian-rhythmic-related species prior to entering the proteinaceous species measurements into the trained rhythmic phase classification model.

49. The method of any one of claims 42-48 further comprising: correcting for individualism using the computational processing system, wherein correcting for individualism comprises: determining a proteinaceous species signature of the individual; determining an individual’s personalized center of the proteinaceous species signature; and adjusting the classification result using the individual’s personalized center based on a difference between a center of a population norm and the individual’s personalized center.

50. The method of any one of claim 42-49 further comprising: measuring proteinaceous species within a proteinaceous species sample, wherein the proteinaceous species sample is or is derived from the biological sample.51 . The method of claim 50, wherein measuring proteinaceous species comprises one or more of the following: ELISA, antibody arrays, DNA bar-coded antibodies, aptamers, chromatography, mass spectrometry, NMR, or electrophoresis.

52. The method of claim 50, wherein measuring proteinaceous species comprises: contacting the proteinaceous species sample with a panel of aptamers that are configured to bind select proteinaceous species; and sequencing the aptamers that bind a proteinaceous species.

53. The method of claim 50, wherein measuring proteinaceous species comprises: contacting the proteinaceous species sample with a panel of DNA bar-coded antibodies that are configured to bind select proteinaceous species; and sequencing DNA bar codes of the antibodies that bind a proteinaceous species.

54. The method of claim 52 or 53, wherein the aptamers or the DNA bar-coded antibodies are configured to bind two or more of the following proteinaceous species: angiopoietin-related protein 1 , cathepsin F, pro-opiomelanocortin, prolactin, metalloproteinase inhibitor 4, hyaluronidase-1 , thyroid stimulating hormone, kallikrein-7, secretogranin-1 , cysteine-rich secretory protein LCCL domain-containing 2, hemopexin, zymogen granule protein 16 homolog B, pancreatic hormone, follistatin-related protein 3, neutrophil cytosol factor 1 , heparan-sulfate 6-O-sulfotransferase 2, cGMP-dependent 3’,5'-cyclic phosphodiesterase, HLA class I alpha chain G, lactotransferrin, proprotein convertase subtilisin / kexin type 9, secreted and transmembrane protein 1 , desmocollin- 2, succinate dehydrogenase assembly factor 2 -mitochondrial, and lysosomal protective protein.

55. The method of any one of claims 42-54 wherein the trained rhythmic phase classification model comprises the use of one or more of: linear regression, polynomial regression, Cox proportional hazards regression, multiple linear regression, ridge regression, logistic regression, Lasso regression, stepwise regression, principal component analysis, Bayesian inference, elastic net, and random forest regression.

Images