Systems and methods for enriching cell-free microbial nucleic acid molecules

Abstract

A system and method are provided for enriching cell-free microbial nucleic acids from one or more target samples.

Description

Technical Field

[0001] Cross-reference This application claims the benefit of U.S. Provisional Application No. 63 / 337,889, filed May 3, 2022, which is hereby incorporated by reference in its entirety for all purposes.

Background Art

[0002] Circulating microbial cell-free nucleic acid molecules have been shown to be promising as disease diagnosis and / or prognosis biomarkers. However, the associated microbial cell-free nucleic acid signatures are often present in small amounts and difficult to detect in the background of mammalian cell-free nucleic acid. Therefore, there are unmet needs for systems and methods to improve the detection of circulating microbial cell-free nucleic acid molecules used for disease diagnosis and / or prognosis.

Summary of the Invention

[0003] The present invention addresses unmet needs by methods and systems configured to deplete mammalian cell-free nucleic acid from a sample, thereby enriching and / or isolating microbial cell-free nucleic acid. In some cases, mammalian cell-free nucleic acid can be depleted via an affinity agent configured to selectively bind to one or more mammalian cell-free nucleic acid molecules. In some examples, one or more mammalian cell-free nucleic acid molecules can be complexed and / or coupled to one or more proteins. In some cases, one or more proteins include histone proteins.

[0004] Aspects of the present disclosure include methods for generating a microbial metagenomic signature set for distinguishing one or more target cancers and non-neoplastic diseases. In some embodiments, the method comprises: (a) providing one or more biological samples from one or more subjects, the samples comprising one or more mammalian nucleic acid molecules and one or more microbial nucleic acid molecules, and a corresponding health state; (b) removing the one or more mammalian nucleic acid molecules from the biological sample using one or more affinity capture reagents; (c) sequencing the remaining one or more microbial nucleic acid molecules to generate one or more microbial sequencing reads; and (d) combining the abundance of one or more metagenomic signatures of the one or more microbial sequencing reads with the health state of the one or more subjects to generate a microbial metagenomic signature set configured to distinguish the presence of cancer or non-cancer disease. In some embodiments, step (b) comprises: (a) contacting the liquid biological sample with a solid support comprising immobilized anti-nucleosome antibodies to form an antibody-nucleosome interaction complex; (b) separating the solid support from the liquid biological sample to concentrate the antibody-nucleosome interaction complex; and (c) purifying the remaining one or more nucleosome-depleted microbial nucleic acid molecules. In some embodiments, the anti-nucleosome antibodies are configured to bind to an epitope comprising DNA and one or more histone proteins. In some embodiments, the solid support comprises magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, pH-sensitive polymers, or any combination thereof.

[0005] In some embodiments, the metagenomic signature set comprises microbial taxonomic abundances. In some embodiments, the metagenomic signature set comprises computationally inferred microbial biochemical pathways and associated abundances of the microbial biochemical pathways. In some embodiments, the metagenomic signature set comprises microbial phylogenetic marker genes or marker gene fragments thereof.

[0006] In some embodiments, step (b) comprises: (a) contacting the liquid biological sample with one or more anti-nucleosome antibodies to form an antibody-nucleosome interaction complex; (b) contacting the antibody-nucleosome interaction complex with a solid support, wherein the surface of the solid support comprises a binding moiety configured to couple to the antibody-nucleosome interaction complex; (c) separating the solid support from the liquid biological sample to concentrate the antibody-nucleosome interaction complex; and (d) purifying the remaining one or more nucleosome-depleted microbial nucleic acid molecules. In some embodiments, the one or more anti-nucleosome antibodies comprise one or more epitope tags. In some embodiments, the one or more epitope tags comprise an N-terminal or C-terminal 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), Fc fusion, biotin, or any combination thereof. In some embodiments, the solid support comprises magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, pH-sensitive polymers, or any combination thereof. In some embodiments, the solid support comprises an affinity agent immobilized covalently. In some embodiments, the affinity reagent comprises streptavidin, 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), an antibody specific for biotin, or any combination thereof. In some embodiments, the affinity agent comprises an anti-species antibody.

[0007] In some embodiments, step (c) includes: (a) generating a single-stranded DNA library from one or more microbial nucleic acid molecules; (b) performing shotgun metagenomic sequencing analysis on the single-stranded DNA library to generate one or more sequencing reads; (c) filtering the one or more sequencing reads to generate one or more mammalian DNA-depleted microbial sequencing reads; and (d) decontaminating the one or more mammalian DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads. In some embodiments, decontamination includes in silico decontamination. In some embodiments, filtering includes computationally mapping the one or more sequencing reads to a human reference genome database.

[0008] In some embodiments, the biological sample includes a liquid biological sample, and the liquid biological sample includes plasma, serum, whole blood, urine, cerebrospinal fluid, saliva, sweat, tears, breath condensate, or any dilutions or processed fractions thereof.

[0009] In some embodiments, step (c) includes: (a) amplifying one or more genomic features of the one or more microbial nucleic acid molecules to generate one or more amplified genomic features; (b) sequencing the one or more amplified genomic features to generate one or more sequencing reads; (c) filtering the one or more sequencing reads to generate one or more mitochondrial DNA-depleted microbial sequencing reads; and (d) decontaminating the one or more mitochondrial DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads. In some embodiments, decontamination includes in silico decontamination. In some embodiments, the one or more genomic features include a microbial phylogenetic marker gene or a marker gene fragment thereof. In some embodiments, the microbial phylogenetic marker gene includes a bacterial marker gene or a marker gene fragment thereof. In some embodiments, the microbial phylogenetic marker gene includes a fungal marker gene or a marker gene fragment thereof.

[0010] In some embodiments, the bacterial marker gene comprises ribosomal RNA gene 5S, ribosomal RNA gene 16S, ribosomal RNA gene 23S, bacterial housekeeping gene dnaG, frr, infC, nusA, pgk, pyrG, rplA, rplB, rplC, rplD, rplE, rplF, rplK, rplL, rplM, rplN, rplP, rplS, rplT, rpmA, rpoB, rpsB, rpsC, rpsE, rpsI, rpsJ, rpsK, rpsM, rpsS, smpB, tsf, or any combination thereof. In some embodiments, the fungal marker gene comprises ribosomal RNA gene 18S, ribosomal RNA gene 5.8S, ribosomal RNA gene 28S, and one or more of internal transcribed spacer regions 1 and 2. In some embodiments, the microbial phylogenetic marker gene comprises marker genes of bacteria, fungi, or any combination thereof. In some embodiments, the amplification comprises performing polymerase chain reaction or a derivative thereof. In some embodiments, the derivative comprises inverse PCR, anchored PCR, primer-directed rolling circle amplification, or any combination thereof.

[0011] In some embodiments, step (c) comprises enriching the one or more microbial nucleic acid molecules. In some embodiments, the enrichment comprises: (a) combining the purified nucleosome-depleted microbial nucleic acid molecules with a hybridization probe, wherein the hybridization probe comprises nucleic acid sequence complementarity to a microbial genomic feature; (b) incubating the hybridization probe and the one or more nucleosome-depleted microbial nucleic acid molecules under conditions that promote nucleic acid base pairing between the target nucleic acid feature and the hybridization probe; (c) separating unbound hybridization probe and hybridized probe bound to the microbial nucleic acid molecule; and (d) washing the hybridized probe bound to the microbial nucleic acid molecule, thereby generating one or more enriched microbial nucleic acid molecules. In some embodiments, the washing removes non-specifically associated nucleic acid molecules and other reaction components.

[0012] In some embodiments, enrichment comprises: (a) combining one or more purified nucleosome-depleted microbial nucleic acid molecules with one or more recombinant CXXC domain proteins to form a protein-DNA binding reaction; (b) incubating the protein-DNA binding reaction under conditions that promote an interaction between the recombinant CXXC domain protein and the unmethylated CpG motifs of the one or more nucleosome-depleted microbial nucleic acid molecules; (c) separating unbound recombinant CXXC domain protein and recombinant CXXC domain protein bound to the unmethylated CpG motifs from the remainder of the protein-DNA binding reaction; (d) washing the recombinant CXXC domain protein bound to the unmethylated CpG nucleic acid fragments, thereby generating one or more enriched nucleic acid molecules for amplification. In some embodiments, the washing is configured to remove non-specifically associated nucleic acid molecules and the remainder of the protein-DNA binding reaction components. In some embodiments, amplification comprises performing polymerase chain reaction or a derivative thereof. In some embodiments, the derivative comprises inverse PCR, anchored PCR, primer-directed rolling circle amplification, or any combination thereof.

[0013] In some embodiments, the one or more subjects include a subject that is a human, a non-human mammal, or any combination thereof. In some embodiments, the one or more mammalian nucleic acid molecules include nucleic acid molecules that are DNA, RNA, cell-free DNA, cell-free RNA, exosomal DNA, exosomal RNA, or any combination thereof, and the one or more microbial nucleic acid molecules include nucleic acid molecules that are microbial cell-free RNA, microbial cell-free DNA, microbial RNA, microbial DNA, or any combination thereof. In some embodiments, the cancer includes acute myeloid leukemia, adrenocortical cancer, bladder urothelial cancer, brain low-grade glioma, breast invasive cancer, cervical squamous cell carcinoma and endocervical adenocarcinoma, cholangiocarcinoma, colon adenocarcinoma, esophageal cancer, glioblastoma multiforme, head and neck squamous cell carcinoma, kidney chromophobe cell, kidney clear cell renal cell carcinoma, kidney papillary renal cell carcinoma, liver hepatocellular carcinoma, lung adenocarcinoma, lung squamous cell carcinoma, lymphoma diffuse large B-cell lymphoma, mesothelioma, ovarian serous cystadenocarcinoma, pancreatic adenocarcinoma, pheochromocytoma and paraganglioma, prostate adenocarcinoma, rectal adenocarcinoma, sarcoma, skin melanoma, stomach adenocarcinoma, testicular germ cell tumor, thymoma, thyroid cancer, uterine carcinosarcoma, endometrial carcinoma of the uterine corpus, choroidal melanoma, or any combination thereof. In some embodiments, the non-neoplastic disease includes a non-cancerous state that is health, disease, or any combination thereof. In some embodiments, the disease state includes a benign tumor of the integumentary, skeletal, muscular, nervous, endocrine, cardiovascular, lymphatic, digestive, respiratory, urinary, reproductive, or any combination of those systems. In some embodiments, the cancer includes stage I, II, or III cancer. In some embodiments, the method further includes generating a trained prediction model, and the trained prediction model is trained with the microbial metagenomic feature set of the one or more subjects and the health state. In some embodiments, the trained prediction model includes a machine learning model, one or more machine learning models, an ensemble of machine learning models, or any combination thereof. In some embodiments, the trained prediction model includes a regularized machine learning model. In some embodiments, the machine learning model includes a machine learning classifier.In some embodiments, the machine learning model includes a gradient boosting machine, a neural network, a support vector machine, k-means, a classification tree, a random forest, regression, or any combination thereof. In some embodiments, the condition of the subject or the subject's health state includes a known non-neoplastic disease of the subject, cancer, or any combination thereof.

[0014] Another aspect of the present disclosure includes a method of using the output of a prediction model trained to diagnose cancer or non-neoplastic diseases of one or more subjects, the method comprising: (a) providing a biological sample of one or more subjects comprising one or more mammalian nucleic acid molecules and one or more microbial nucleic acid molecules; (b) removing the one or more mammalian nucleic acid molecules from the biological sample using one or more affinity capture reagents; (c) sequencing the remaining one or more microbial nucleic acid molecules to generate one or more microbial sequencing reads; (d) generating one or more microbial metagenomic feature sets by combining the abundances of one or more metagenomic features of the one or more microbial sequencing reads; and (e) outputting a diagnosis of cancer or non-neoplastic diseases of the one or more subjects, at least as a result of providing the one or more microbial metagenomic feature sets as an input to a trained prediction model. In some embodiments, the one or more microbial metagenomic feature sets include microbial taxonomic abundances. In some embodiments, the one or more microbial metagenomic feature sets include computationally inferred microbial biochemical pathways and their associated abundances. In some embodiments, the one or more microbial metagenomic feature sets include microbial phylogenetic marker genes or marker gene fragments thereof.

[0015] In some embodiments, the biological sample includes a liquid biological sample, and the liquid biological sample includes plasma, serum, whole blood, urine, cerebrospinal fluid, saliva, sweat, tears, exhaled condensate, or any combination, dilution, or processed fraction thereof.

[0016] In some embodiments, step (b) comprises: (a) contacting the liquid biological sample with a solid support comprising an immobilized anti-nucleosome antibody, wherein the anti-nucleosome antibody is configured to form an antibody-nucleosome interaction complex; (b) separating the solid support from the liquid biological sample to concentrate the antibody-nucleosome interaction complex; and (c) purifying the remaining one or more nucleosome-depleted microbial nucleic acid molecules. In some embodiments, the anti-nucleosome antibody recognizes an epitope comprising DNA and one or more histone proteins. In some embodiments, the solid support can comprise magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, pH-sensitive polymers, or any combination thereof.

[0017] In some embodiments, step (b) comprises: (a) contacting the liquid biological sample with one or more anti-nucleosome antibodies to form an antibody-nucleosome interaction complex; (b) contacting the antibody-nucleosome interaction complex with a solid support configured to bind to the antibody-nucleosome interaction complex; (c) separating the solid support from the liquid biological sample to concentrate the antibody-nucleosome interaction complex; and (d) purifying the remaining one or more nucleosome-depleted microbial nucleic acids. In some embodiments, the one or more anti-nucleosome antibodies comprise one or more epitope tags. In some embodiments, the one or more epitope tags comprise an N-terminal or C-terminal 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), Fc fusion, biotin, or any combination thereof. In some embodiments, the solid support comprises magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, pH-sensitive polymer, or any combination thereof. In some embodiments, the solid support comprises an affinity agent immobilized by covalent bond. In some embodiments, the affinity agent immobilized by covalent bond comprises streptavidin, 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), an antibody specific for biotin, or any combination thereof. In some embodiments, the affinity agent immobilized by covalent bond comprises an anti-species antibody.

[0018] In some embodiments, step (c) comprises: (a) generating a single-stranded DNA library from the one or more microbial nucleic acid molecules; (b) performing shotgun metagenomic sequencing analysis of the single-stranded DNA library to generate one or more sequencing reads; (c) filtering the one or more sequencing reads to generate one or more mammalian DNA-depleted microbial sequencing reads; and (d) purifying the one or more mammalian DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads. In some embodiments, purification comprises in silico purification of the one or more mammalian DNA-depleted microbial sequencing reads. In some embodiments, filtering comprises computationally mapping the one or more sequencing reads to a human reference genome database.

[0019] In some embodiments, step (c) includes: (a) amplifying one or more genomic features of the one or more microbial nucleic acid molecules, thereby generating one or more amplified genomic features; (b) sequencing the one or more amplified genomic features to generate one or more sequencing reads; (c) filtering the one or more sequencing reads to produce one or more mitochondrial DNA-depleted microbial sequencing reads; and (d) purifying the one or more mitochondrial DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads. In some embodiments, purification includes in silico purification of the one or more mitochondrial DNA-depleted microbial sequencing reads. In some embodiments, the one or more genomic features include a microbial phylogenetic marker gene or a marker gene fragment thereof. In some embodiments, the microbial phylogenetic marker gene includes a bacterial marker gene or a marker gene fragment thereof. In some embodiments, the microbial phylogenetic marker gene includes a fungal marker gene or a marker gene fragment thereof. In some embodiments, the bacterial marker gene includes ribosomal RNA gene 5S, ribosomal RNA gene 16S, ribosomal RNA gene 23S, bacterial housekeeping gene dnaG, frr, infC, nusA, pgk, pyrG, rplA, rplB, rplC, rplD, rplE, rplF, rplK, rplL, rplM, rplN, rplP, rplS, rplT, rpmA, rpoB, rpsB, rpsC, rpsE, rpsI, rpsJ, rpsK, rpsM, rpsS, smpB, tsf, or any combination thereof. In some embodiments, the fungal marker gene includes ribosomal RNA gene 18S, ribosomal RNA gene 5.8S, ribosomal RNA gene 28S, and one or more of internal transcribed spacer regions 1 and 2. In some embodiments, the microbial phylogenetic marker gene includes marker genes of bacteria, fungi, or any combination thereof. In some embodiments, amplification includes performing polymerase chain reaction or a derivative thereof.In some embodiments, the derivatives thereof include reverse PCR, anchor PCR, primer-directed rolling circle amplification, or any combination thereof.

[0020] In some embodiments, step (c) includes enriching the one or more microbial nucleic acid molecules. In some embodiments, the enrichment of the one or more microbial nucleic acid molecules comprises: (a) combining the purified one or more nucleosome-depleted microbial nucleic acid molecules with a hybridization probe, wherein the hybridization probe comprises a nucleic acid sequence complementary to a microbial genomic nucleic acid feature; (b) incubating the hybridization probe and the one or more nucleosome-depleted microbial nucleic acid molecules under conditions that promote nucleic acid base pairing between the microbial genomic nucleic acid feature and the hybridization probe; (c) separating unbound hybridization probe and hybridized probes bound to the one or more nucleosome-depleted microbial nucleic acid molecules; and (d) washing the hybridized probes bound to the one or more nucleosome-depleted microbial nucleic acid molecules, thereby generating one or more enriched microbial nucleic acid molecules. In some embodiments, the washing is configured to remove non-specifically associated nucleic acid molecules and other reaction components.

[0021] In some embodiments, the enrichment of the one or more microbial nucleic acid molecules comprises: (a) combining one or more purified nucleosome-depleted microbial nucleic acid molecules with one or more recombinant CXXC domain proteins to form a protein-DNA binding reaction; (b) incubating the protein-DNA binding reaction under conditions that promote the interaction between the recombinant CXXC domain protein and the non-methylated CpG motifs of the one or more nucleosome-depleted microbial nucleic acid molecules; (c) separating the unbound recombinant CXXC domain protein and the recombinant CXXC domain protein bound to the non-methylated CpG nucleic acid fragment from the remainder of the protein-DNA binding reaction components; (d) washing the recombinant CXXC domain protein bound to the non-methylated CpG nucleic acid fragment, thereby generating one or more enriched nucleic acid molecules for amplification. In some embodiments, the washing is configured to remove non-specifically associated nucleic acid molecules and the remainder of the protein-DNA binding reaction components. In some embodiments, the amplification comprises performing a polymerase chain reaction or a derivative thereof. In some embodiments, the derivative comprises inverse PCR, anchored PCR, primer-directed rolling circle amplification, or any combination thereof. In some embodiments, the one or more mammalian nucleic acid molecules and the one or more microbial nucleic acid molecules are derived from one or more liquid biological samples of the one or more subjects. In some embodiments, the one or more subjects include human, non-human mammalian, or any combination thereof. In some embodiments, the one or more mammalian nucleic acid molecules include nucleic acid molecules of DNA, RNA, cell-free RNA, cell-free DNA, exosomal DNA, exosomal RNA, or any combination thereof, and the one or more microbial nucleic acid molecules include nucleic acid molecules of microbial cell-free DNA, microbial cell-free RNA, microbial DNA, microbial RNA, or any combination thereof.In some embodiments, the cancer includes acute myeloid leukemia, adrenocortical carcinoma, bladder urothelial carcinoma, low-grade glioma of the brain, invasive breast cancer, squamous cell carcinoma and adenocarcinoma of the cervix, cholangiocarcinoma, colorectal adenocarcinoma, esophageal cancer, glioblastoma multiforme, squamous cell carcinoma of the head and neck, clear cell renal cell carcinoma, papillary renal cell carcinoma of the kidney, hepatocellular carcinoma of the liver, lung adenocarcinoma, lung squamous cell carcinoma, diffuse large B-cell lymphoma, mesothelioma, serous cystadenocarcinoma of the ovary, pancreatic adenocarcinoma, pheochromocytoma and paraganglioma, prostate adenocarcinoma, rectal adenocarcinoma, sarcoma, cutaneous melanoma, gastric adenocarcinoma, testicular germ cell tumor, thymoma, thyroid cancer, uterine carcinosarcoma, endometrial cancer of the uterine corpus, choroidal melanoma, or any combination thereof. In some embodiments, the non-neoplastic disease includes a non-cancerous state of health, disease, or any combination thereof. In some embodiments, the disease state includes a benign tumor of the integumentary, skeletal, muscular, nervous, endocrine, cardiovascular, lymphatic, digestive, respiratory, urinary, reproductive, or any combination of those systems. In some embodiments, the cancer includes stage I, II, or III cancer.

[0022] In some embodiments, the trained prediction model is trained with one or more microbial metagenomic feature sets of one or more subjects and the corresponding health states. In some embodiments, the trained prediction model includes a machine learning model, one or more machine learning models, an ensemble of machine learning models, or any combination thereof. In some embodiments, the trained prediction model includes a regularized machine learning model. In some embodiments, the machine learning model includes a machine learning classifier. In some embodiments, the machine learning model includes a gradient boosting machine, neural network, support vector machine, k-means, classification tree, random forest, regression, or any combination thereof of machine learning models. In some embodiments, the subject's or the subject's health state includes a known non-neoplastic disease, cancer, or any combination thereof of the subject.

[0023] Another aspect of the present disclosure includes a system for diagnosing a cancerous or non-cancerous health state of one or more subjects. In some embodiments, the system includes (a) a processor and (b) software configured to cause the processor to (i) receive sequence determination reads of mammalian nucleosome-depleted nucleic acid molecules of the one or more subjects from a liquid biological sample of the one or more subjects, wherein the one or more nucleic acid molecule sequence determination reads include sequences of one or more metagenomic features of one or more microbial nucleic acid molecules, and (ii) output a diagnosis of the cancerous or non-cancerous health state of the one or more subjects as a result of providing at least the one or more metagenomic features of the one or more microbial nucleic acid sequence determination reads as an input to a trained prediction model. In some embodiments, the one or more metagenomic features include microbial taxonomic abundances. In some embodiments, the one or more metagenomic features include computationally inferred microbial biochemical pathways and their associated abundances. In some embodiments, the one or more metagenomic features include microbial phylogenetic marker genes or marker gene fragments thereof. In some embodiments, the mammalian nucleosome-depleted nucleic acid molecule sequence determination reads are obtained from and / or received from a liquid biological sample of the subject, and the liquid biological sample includes plasma, serum, whole blood, urine, cerebrospinal fluid, saliva, sweat, tears, exhaled condensate, or any combination, dilution, or processed fraction thereof.

[0024] In some embodiments, sequencing reads of one or more mammalian nucleosome-depleted nucleic acid molecules are generated by: (a) contacting the liquid biological sample with a solid support to form an antibody-nucleosome interaction complex, wherein the solid support comprises a surface having an anti-nucleosome antibody coupled thereto; (b) separating the solid support from the liquid biological sample to concentrate the antibody-nucleosome interaction complex; (c) purifying the remaining one or more nucleosome-depleted microbial nucleic acid molecules; and (d) sequencing the purified one or more nucleosome-depleted microbial nucleic acid molecules. In some embodiments, the anti-nucleosome antibody is configured to recognize an epitope comprising DNA and one or more histone proteins. In some embodiments, the solid support comprises magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, pH-sensitive polymers, or any combination thereof.

[0025] In some embodiments, the sequencing reads of one or more mammalian nucleosome-depleted nucleic acid molecules are generated by: (a) contacting the liquid biological sample with one or more anti-nucleosome antibodies to form an antibody-nucleosome interaction complex; (b) contacting the antibody-nucleosome interaction complex with a solid support; (c) separating the solid support from the liquid biological sample to concentrate the antibody-nucleosome interaction complex; (d) purifying the remaining one or more nucleosome-depleted microbial nucleic acid molecules; and (e) sequencing the purified one or more nucleosome-depleted microbial nucleic acid molecules. In some embodiments, the one or more anti-nucleosome antibodies comprise one or more epitope tags. In some embodiments, the one or more epitope tags comprise an N-terminal or C-terminal 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), an Fc fusion, biotin, or any combination thereof. In some embodiments, the solid support comprises magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, a pH-sensitive polymer, or any combination thereof. In some embodiments, the solid support comprises an affinity agent immobilized by covalent bonding. In some embodiments, the affinity agent immobilized by covalent bonding comprises streptavidin, a 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), an antibody specific for biotin, or any combination thereof. In some embodiments, the affinity agent immobilized by covalent bonding comprises an anti-species antibody.

[0026] In some embodiments, sequencing reads of one or more mammalian nucleosome-depleted nucleic acid molecules are generated by: (a) generating a single-stranded DNA library from the one or more microbial nucleic acid molecules; (b) performing shotgun metagenomic sequencing analysis of the single-stranded DNA library to generate one or more sequencing reads; (c) filtering the one or more sequencing reads to generate one or more mammalian DNA-depleted microbial sequencing reads; and (d) purifying the one or more mammalian DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads. In some embodiments, purification includes in silico purification of the one or more mammalian DNA-depleted microbial sequencing reads. In some embodiments, filtering includes computationally mapping the one or more sequencing reads to a human reference genome database.

[0027] In some embodiments, mammalian nucleosome-depleted nucleic acid molecule sequencing reads are generated by: (a) amplifying one or more genomic features of the one or more microbial nucleic acid molecules, thereby generating one or more amplified genomic features; (b) sequencing the one or more amplified genomic features to generate one or more sequencing reads; (c) filtering the one or more sequencing reads to generate one or more mitochondrial DNA-depleted microbial sequencing reads; and (d) purifying the one or more mitochondrial DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads. In some embodiments, purification includes in silico purification of the one or more mitochondrial DNA-depleted microbial sequencing reads. In some embodiments, the one or more genomic features include a microbial phylogenetic marker gene or a marker gene fragment thereof. In some embodiments, the microbial phylogenetic marker gene includes a bacterial marker gene or a marker gene fragment thereof. In some embodiments, the microbial phylogenetic marker gene includes a fungal marker gene or a marker gene fragment thereof. In some embodiments, the bacterial marker gene includes ribosomal RNA gene 5S, ribosomal RNA gene 16S, ribosomal RNA gene 23S, bacterial housekeeping gene dnaG, frr, infC, nusA, pgk, pyrG, rplA, rplB, rplC, rplD, rplE, rplF, rplK, rplL, rplM, rplN, rplP, rplS, rplT, rpmA, rpoB, rpsB, rpsC, rpsE, rpsI, rpsJ, rpsK, rpsM, rpsS, smpB, tsf, or any combination thereof. In some embodiments, the fungal marker gene includes ribosomal RNA gene 18S, ribosomal RNA gene 5.8S, ribosomal RNA gene 28S, and one or more of internal transcribed spacer regions 1 and 2. In some embodiments, the microbial phylogenetic marker gene includes marker genes of bacteria, fungi, or any combination thereof. In some embodiments, amplification includes performing polymerase chain reaction (PCR) or a derivative thereof.In some embodiments, the derivatives include reverse PCR, anchored PCR, primer-directed rolling circle amplification, or any combination thereof. In some embodiments, the one or more microbial nucleic acid molecules are enriched from the one or more mammalian nucleosome-depleted nucleic acid molecules.

[0028] In some embodiments, enrichment of the one or more microbial nucleic acid molecules comprises: (a) combining a purified nucleosome-depleted microbial nucleic acid molecule with a hybridization probe, the hybridization probe comprising nucleic acid sequence complementarity to one or more microbial genomic nucleic acid features; (b) incubating the hybridization probe and the one or more nucleosome-depleted microbial nucleic acid molecules under conditions that promote nucleic acid base pairing between the one or more microbial genomic nucleic acid features and the hybridization probe; (c) separating unbound hybridization probe and hybridized probe bound to the one or more nucleosome-depleted microbial nucleic acid molecules; and (d) washing the hybridized probe bound to the one or more nucleosome-depleted microbial nucleic acid molecules, thereby generating one or more enriched microbial nucleic acid molecules. In some embodiments, the washing is configured to remove non-specifically associated nucleic acid molecules and other reaction components.

[0029] In some embodiments, the enrichment of the one or more microbial nucleic acid molecules comprises: (a) combining one or more purified nucleosome-depleted microbial nucleic acid molecules with one or more recombinant CXXC domain proteins to form a protein-DNA binding reaction; (b) incubating the protein-DNA binding reaction under conditions that promote the interaction between the recombinant CXXC domain protein and the unmethylated CpG motifs of the one or more nucleosome-depleted microbial nucleic acid molecules; (c) separating unbound recombinant CXXC domain proteins and recombinant CXXC domain proteins bound to the unmethylated CpG nucleic acid fragments from the remainder of the protein-DNA binding reaction components; (d) washing the recombinant CXXC domain proteins bound to the unmethylated CpG nucleic acid fragments, thereby generating one or more enriched nucleic acid molecules for amplification. In some embodiments, the washing is configured to remove nonspecifically associated nucleic acid molecules and the remainder of the protein-DNA binding reaction components. In some embodiments, the amplification comprises performing a polymerase chain reaction (PCR) or a derivative thereof. In some embodiments, the derivative comprises inverse PCR, anchor PCR, primer-directed rolling circle amplification, or any combination thereof. In some embodiments, the one or more mammalian nucleic acid molecules and the one or more microbial nucleic acid molecules are derived from one or more liquid biological samples of the one or more subjects. In some embodiments, the one or more subjects include human, non-human mammalian, or any combination thereof. In some embodiments, mammalian nucleosome-depleted nucleic acid molecule sequencing reads are obtained from mammalian nucleosome-depleted nucleic acid molecules of a biological sample of a subject, the biological sample comprising one or more mammalian nucleic acid molecules including DNA, RNA, cell-free RNA, cell-free DNA, exosomal DNA, exosomal RNA, or any combination thereof, and the one or more microbial nucleic acid molecules comprising microbial cell-free DNA, microbial cell-free RNA, microbial DNA, microbial RNA, or any combination thereof.In some embodiments, the cancer includes acute myeloid leukemia, adrenocortical carcinoma, bladder urothelial carcinoma, low-grade glioma of the brain, invasive breast cancer, squamous cell carcinoma and adenocarcinoma of the cervix, cholangiocarcinoma, colorectal adenocarcinoma, esophageal cancer, glioblastoma multiforme, squamous cell carcinoma of the head and neck, clear cell of the kidney, renal clear cell carcinoma, renal papillary cell carcinoma of the kidney, hepatocellular carcinoma of the liver, lung adenocarcinoma, lung squamous cell carcinoma, lymphoma diffuse large B-cell lymphoma, mesothelioma, serous cystadenocarcinoma of the ovary, pancreatic adenocarcinoma, pheochromocytoma and paraganglioma, prostate adenocarcinoma, rectal adenocarcinoma, sarcoma, cutaneous melanoma, gastric adenocarcinoma, testicular germ cell tumor, thymoma, thyroid cancer, uterine carcinosarcoma, endometrial carcinoma of the uterine body, choroidal melanoma, or any combination thereof. In some embodiments, the non-neoplastic disease includes a non-cancerous state of health, disease, or any combination thereof. In some embodiments, the disease state includes a benign tumor of the integumentary, skeletal, muscular, nervous, endocrine, cardiovascular, lymphatic, digestive, respiratory, urinary, reproductive, or any combination of those systems.

[0030] In some embodiments, the cancer includes stage I, II, or III cancer.

[0031] In some embodiments, the trained prediction model is trained with one or more metagenomic features of one or more subjects and the corresponding health states. In some embodiments, the trained prediction model includes a machine learning model, one or more machine learning models, an ensemble of machine learning models, or any combination thereof. In some embodiments, the trained prediction model includes a regularized machine learning model. In some embodiments, the machine learning model includes a machine learning classifier. In some embodiments, the machine learning model includes a gradient boosting machine, a neural network, a support vector machine, k-means, a classification tree, a random forest, regression, or any combination thereof of machine learning models. In some embodiments, the subject's or subject's health state includes a known non-neoplastic disease, cancer, or any combination thereof of the subject.

[0032] Another aspect of the present disclosure includes a method for enriching cell-free microbial nucleic acid molecules of a sample. In some embodiments, the method comprises: (a) contacting a sample of one or more cell-free nucleic acid molecules with a first set of one or more probes, wherein the first set of one or more probes comprises a binding moiety configured to bind to one or more human nucleic acid molecules complexed with one or more proteins; and (b) enriching one or more cell-free microbial nucleic acid molecules of the sample by removing the one or more probes bound to the one or more human nucleic acid molecules complexed with the one or more proteins from the sample. In some embodiments, the one or more proteins include one or more histone proteins, one or more regulatory proteins, or any combination thereof. In some embodiments, the sample includes plasma, serum, whole blood, urine, cerebrospinal fluid, saliva, sweat, tears, exhaled condensate, or any combination, dilution, or processed fraction thereof. In some embodiments, the one or more probes include one or more antibodies. In some embodiments, the removing comprises incubating the one or more antibodies bound to the one or more human nucleic acid molecules complexed with the one or more proteins with a solid support, the solid support including one or more capture reagents configured to bind to the one or more antibodies.

[0033] In some embodiments, the method further comprises (c) contacting the enriched cell-free microbial nucleic acid molecules with a second set of one or more probes, wherein the second set of one or more probes is configured to bind to one or more microbial marker genes. In some embodiments, the one or more microbial marker genes include ribosomal RNA gene 5S, ribosomal RNA gene 16S, ribosomal RNA gene 23S, bacterial housekeeping gene dnaG, frr, infC, nusA, pgk, pyrG, rplA, rplB, rplC, rplD, rplE, rplF, rplK, rplL, rplM, rplN, rplP, rplS, rplT, rpmA, rpoB, rpsB, rpsC, rpsE, rpsI, rpsJ, rpsK, rpsM, rpsS, smpB, tsf, or any combination thereof. In some embodiments, the one or more microbial marker genes are sequenced to determine the abundance of the microbial taxonomic, functional, or any combination thereof. In some embodiments, the sample includes a liquid biological sample. In some embodiments, the sample is derived from a subject. In some embodiments, the subject is a human or non-human mammal. In some embodiments, the one or more proteins include histone proteins associated with the one or more nucleic acid molecules. In some embodiments, the one or more human nucleic acid molecules include DNA, RNA, cell-free RNA, cell-free DNA, exosomal RNA, exosomal DNA, or any combination thereof. In some embodiments, the one or more cell-free microbial nucleic acid molecules include cell-free microbial DNA, cell-free microbial RNA, microbial RNA, microbial DNA, or any combination thereof. In some embodiments, the removal comprises immunoprecipitating the one or more probes bound to the one or more human nucleic acid molecules.

[0034] In some embodiments, the method further comprises (d) preparing a single-stranded library from the one or more cell-free microbial nucleic acid molecules of the sample. In some embodiments, the first set of one or more probes is coupled to a solid support. In some embodiments, the solid support comprises beads, magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, pH-sensitive polymers, or any combination thereof. In some embodiments, the sample comprises one or more human nucleic acid molecules, one or more microbial nucleic acid molecules, or a combination thereof. In some embodiments, the health condition of the subject or subjects comprises a known non-neoplastic disease of the subject, cancer, or any combination thereof.

[0035] Aspects of the present disclosure include a method for generating a microbial metagenomic signature set for diagnosing a disease, the method comprising: (a) providing the health conditions of a plurality of subjects and biological samples of the plurality of subjects, wherein the biological samples comprise mammalian nucleic acid molecules and microbial nucleic acid molecules; (b) removing the mammalian nucleic acid molecules from the biological samples using an affinity capture reagent; (c) sequencing the microbial nucleic acid molecules to generate microbial sequencing reads; and (d) generating the microbial metagenomic signature set for diagnosing the disease by combining the metagenomic signature abundances of the microbial sequencing reads with the health conditions of the plurality of subjects. In some embodiments, the metagenomic signature set comprises microbial taxonomic abundances. In some embodiments, the metagenomic signature set comprises computationally inferred microbial biochemical pathways and associated abundances of the microbial biochemical pathways. In some embodiments, the metagenomic signature set comprises microbial phylogenetic marker genes or marker gene fragments thereof.

[0036] In some embodiments, the biological sample comprises a liquid biological sample, and the liquid biological sample comprises plasma, serum, whole blood, urine, cerebrospinal fluid, saliva, sweat, tears, breath condensate, or any combination, dilution, or processed fraction thereof.

[0037] In some embodiments, step (b) of the method for generating a microbial metagenomic signature set for diagnosing the disease comprises: (a) contacting the liquid biological sample with a solid support comprising an immobilized anti-nucleosome antibody to form an antibody-nucleosome interaction complex; (b) separating the solid support from the liquid biological sample to concentrate the antibody-nucleosome interaction complex; and (c) purifying the remaining one or more nucleosome-depleted microbial nucleic acid molecules. In some embodiments, the anti-nucleosome antibody is configured to bind to an epitope comprising DNA and one or more histone proteins. In some embodiments, the solid support comprises magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, pH-sensitive polymers, or any combination thereof.

[0038] In some embodiments, step (b) of the method of generating a microbial metagenomic signature set for diagnosing the disease comprises: (a) contacting the liquid biological sample with one or more anti-nucleosome antibodies to form an antibody-nucleosome interaction complex; (b) contacting the antibody-nucleosome interaction complex with a solid support; (c) separating the solid support from the liquid biological sample to concentrate the antibody-nucleosome interaction complex; and (d) purifying the remaining nucleosome-depleted microbial nucleic acid molecules. In some embodiments, the anti-nucleosome antibody comprises a plurality of epitope tags. In some embodiments, the plurality of epitope tags comprises an N-terminal or C-terminal 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), an Fc fusion, biotin, or any combination thereof. In some embodiments, the solid support comprises magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, a pH-sensitive polymer, or any combination thereof. In some embodiments, the solid support comprises an affinity agent immobilized by covalent bonding. In some embodiments, the affinity reagent comprises streptavidin, a 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), an antibody specific for biotin, or any combination thereof. In some embodiments, the affinity agent comprises an anti-species antibody.

[0039] In some embodiments, step (c) of the method of generating a microbial metagenomic feature set for diagnosing the disease comprises: (a) generating a single-stranded DNA library from the microbial nucleic acid molecules; (b) performing shotgun metagenomic sequencing analysis of the single-stranded DNA library to generate sequencing reads; (c) filtering the sequencing reads to generate mammalian DNA-depleted microbial sequencing reads; and (d) purifying the mammalian DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads. In some embodiments, the purification includes in silico purification. In some embodiments, the filtering includes computationally mapping the sequencing reads to a human reference genome database.

[0040] In some embodiments, step (c) of the method of generating a microbial metagenomic feature set for diagnosing the disease comprises: (a) amplifying genomic features of the microbial nucleic acid molecules, thereby generating amplified genomic features; (b) sequencing the amplified genomic features to generate sequencing reads; (c) filtering the sequencing reads to generate mitochondrial DNA-depleted microbial sequencing reads; and (d) purifying the mitochondrial DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads. In some embodiments, the purification includes in silico purification. In some embodiments, the genomic features include microbial phylogenetic marker genes or marker gene fragments thereof.

[0041] In some embodiments, the microbial phylogenetic marker gene comprises a bacterial marker gene or a marker gene fragment thereof. In some embodiments, the microbial phylogenetic marker gene comprises a fungal marker gene or a marker gene fragment thereof. In some embodiments, the bacterial marker gene comprises ribosomal RNA gene 5S, ribosomal RNA gene 16S, ribosomal RNA gene 23S, bacterial housekeeping gene dnaG, frr, infC, nusA, pgk, pyrG, rplA, rplB, rplC, rplD, rplE, rplF, rplK, rplL, rplM, rplN, rplP, rplS, rplT, rpmA, rpoB, rpsB, rpsC, rpsE, rpsI, rpsJ, rpsK, rpsM, rpsS, smpB, tsf, or any combination thereof. In some embodiments, the fungal marker gene comprises ribosomal RNA gene 18S, ribosomal RNA gene 5.8S, ribosomal RNA gene 28S, and one or more of internal transcribed spacer regions 1 and 2. In some embodiments, the microbial phylogenetic marker gene comprises a marker gene of bacteria, fungi, or any combination thereof. In some embodiments, the amplification comprises performing polymerase chain reaction or a derivative thereof. In some embodiments, the derivative comprises inverse PCR, anchor PCR, primer-directed rolling circle amplification, or any combination thereof.

[0042] In some embodiments, step (c) of the method of generating a microbial metagenomic feature set for diagnosing the disease comprises enriching the microbial nucleic acid molecules. In some embodiments, the enrichment comprises: (a) combining the microbial nucleic acid molecules with hybridization probes, wherein the hybridization probes comprise nucleic acid sequence complementarity to microbial genomic features; (b) incubating the hybridization probes and the microbial nucleic acid molecules under conditions that promote nucleic acid base pairing between the target nucleic acid features and the hybridization probes; (c) separating unbound hybridization probes and hybridized probes bound to the microbial nucleic acid molecules; and (d) washing the hybridized probes bound to the microbial nucleic acid molecules, thereby generating enriched microbial nucleic acid molecules. In some embodiments, the washing is configured to remove non-specifically associated nucleic acid molecules and other reaction components. In some embodiments, the enrichment comprises: (a) combining the microbial nucleic acid molecules with a recombinant CXXC domain protein to form a protein-DNA binding reaction; (b) incubating the protein-DNA binding reaction under conditions that promote interaction between the recombinant CXXC domain protein and the unmethylated CpG motifs of the microbial nucleic acid molecules; (c) separating unbound recombinant CXXC domain protein and recombinant CXXC domain protein bound to the unmethylated CpG motifs from the remainder of the protein-DNA binding reaction; and (d) washing the recombinant CXXC domain protein bound to the unmethylated CpG nucleic acid fragments, thereby generating enriched nucleic acid molecules for amplification. In some embodiments, the washing is configured to remove non-specifically associated nucleic acid molecules and the remainder of the protein-DNA binding reaction components. In some embodiments, the amplification comprises performing polymerase chain reaction or a derivative thereof. In some embodiments, the derivative comprises inverse PCR, anchor PCR, primer-directed rolling circle amplification, or any combination thereof.In some embodiments, the mammalian nucleic acid molecules and the microbial nucleic acid molecules are derived from the liquid biological samples of the plurality of subjects. In some embodiments, the plurality of subjects includes subjects that are human, non-human mammals, or any combination thereof.

[0043] In some embodiments, the mammalian nucleic acid molecules include nucleic acid molecules that are DNA, RNA, cell-free DNA, cell-free RNA, exosomal DNA, exosomal RNA, or any combination thereof, and the microbial nucleic acid molecules include nucleic acid molecules that are microbial cell-free RNA, microbial cell-free DNA, microbial RNA, microbial DNA, or any combination thereof.

[0044] In some embodiments, the method further includes generating a trained prediction model, where the trained prediction model is trained with the microbial metagenomic feature set of the one or more of the plurality of subjects and the health state. In some embodiments, the trained prediction model includes a machine learning model, one or more machine learning models, an ensemble of machine learning models, or any combination thereof. In some embodiments, the trained prediction model includes a regularized machine learning model. In some embodiments, the machine learning model includes a machine learning classifier. In some embodiments, the machine learning model includes a gradient boosting machine, a neural network, a support vector machine, k-means, a classification tree, a random forest, regression, or any combination thereof of machine learning models. In some embodiments, the health state of the subject or subjects includes a known non-neoplastic disease, cancer, or any combination thereof of the subject.

[0045] Aspects of the present disclosure include a method for diagnosing a disease in a subject, the method comprising: (a) providing a liquid biological sample of the subject, wherein the liquid biological sample comprises mammalian nucleic acid molecules and microbial nucleic acid molecules; (b) removing the mammalian nucleic acid molecules from the liquid biological sample using an affinity capture reagent; (c) sequencing a plurality of microbial nucleic acid molecules of the liquid biological sample to generate microbial sequencing reads; (d) generating a metagenomic feature abundance of the microbial sequencing reads; and (e) outputting a diagnosis of the disease in the subject as a result of providing at least the metagenomic feature abundance as an input to a trained prediction model. In some embodiments, the disease includes a benign tumor of the integumentary, skeletal, muscular, nervous, endocrine, cardiovascular, lymphatic, digestive, respiratory, urinary, reproductive, or any combination of these systems.

[0046] Aspects of the present disclosure include a system for diagnosing a disease in a subject, the system comprising: (a) a processor; and (b) a non-transitory computer-readable storage medium including software configured to cause the processor to: (i) receive a mammalian nucleosome-depleted nucleic acid molecule sequencing read of the subject, wherein the mammalian nucleosome-depleted nucleic acid molecule sequencing read comprises metagenomic features of microbial nucleic acid molecules; and (ii) output a diagnosis of the disease in the subject as a result of providing at least the metagenomic features as an input to a trained prediction model. In some embodiments, the disease includes a benign tumor of the integumentary, skeletal, muscular, nervous, endocrine, cardiovascular, lymphatic, digestive, respiratory, urinary, reproductive, or any combination of these systems. In some embodiments, the subject's or the subject's health status includes a known non-tumorous disease, cancer, or any combination of these of the subject.

[0047] Aspects of the present disclosure include a method for generating a microbial metagenomic feature set for diagnosing cancer, the method comprising: (a) providing the health status of a plurality of subjects and a liquid biological sample of the plurality of subjects, wherein the liquid biological sample comprises mammalian nucleic acid molecules and microbial nucleic acid molecules; (b) removing the mammalian nucleic acid molecules from the liquid biological sample using an affinity capture reagent; (c) sequencing the microbial nucleic acid molecules to generate microbial sequencing reads; and (d) generating the microbial metagenomic feature set for diagnosing the cancer by combining the metagenomic feature abundance of the microbial sequencing reads with the health status of the plurality of subjects. In some embodiments, the metagenomic feature set includes microbial taxonomic abundances. In some embodiments, the metagenomic feature set includes computationally inferred microbial biochemical pathways and associated abundances of the microbial biochemical pathways. In some embodiments, the metagenomic feature set includes microbial phylogenetic marker genes or marker gene fragments thereof. In some embodiments, the liquid biological sample includes plasma, serum, whole blood, urine, cerebrospinal fluid, saliva, sweat, tears, exhaled condensate, or any combination, dilution, or processed fraction thereof.

[0048] In some embodiments, step (b) of the method for generating a microbial metagenomic feature set for diagnosing cancer comprises: (a) contacting the liquid biological sample with a solid support comprising an immobilized anti-nucleosome antibody to form an antibody-nucleosome interaction complex; (b) separating the solid support from the liquid biological sample to concentrate the antibody-nucleosome interaction complex; and (c) purifying the remaining one or more nucleosome-depleted microbial nucleic acid molecules. In some embodiments, the anti-nucleosome antibody is configured to bind to an epitope comprising DNA and one or more histone proteins. In some embodiments, the solid support includes magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, pH-sensitive polymers, or any combination thereof.

[0049] In some embodiments, step (b) of the method for generating a microbial metagenomic signature set for diagnosing cancer comprises: (a) contacting the liquid biological sample with one or more anti-nucleosome antibodies to form an antibody-nucleosome interaction complex; (b) contacting the antibody-nucleosome interaction complex with a solid support; (c) separating the solid support from the liquid biological sample to concentrate the antibody-nucleosome interaction complex; and (d) purifying the remaining nucleosome-depleted microbial nucleic acid molecules. In some embodiments, the anti-nucleosome antibody comprises a plurality of epitope tags. In some embodiments, the plurality of epitope tags comprises an N-terminal or C-terminal 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), an Fc fusion, biotin, or any combination thereof. In some embodiments, the solid support comprises magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, a pH-sensitive polymer, or any combination thereof. In some embodiments, the solid support comprises an affinity agent immobilized by covalent bonding. In some embodiments, the affinity reagent comprises streptavidin, a 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), an antibody specific for biotin, or any combination thereof. In some embodiments, the affinity agent comprises an anti-species antibody.

[0050] In some embodiments, step (c) of the method for generating a microbial metagenomic signature set for diagnosing cancer includes: (a) generating a single-stranded DNA library from the microbial nucleic acid molecules; (b) performing shotgun metagenomic sequencing analysis of the single-stranded DNA library to generate sequencing reads; (c) filtering the sequencing reads to generate mammalian DNA-depleted microbial sequencing reads; and (d) purifying the mammalian DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads. In some embodiments, the purification includes in silico purification. In some embodiments, the filtering includes computationally mapping the sequencing reads to a human reference genome database.

[0051] In some embodiments, step (c) of the method for generating a microbial metagenomic signature set for diagnosing cancer includes: (a) amplifying genomic signatures of the microbial nucleic acid molecules, thereby generating amplified genomic signatures; (b) sequencing the amplified genomic signatures to generate sequencing reads; (c) filtering the sequencing reads to produce mitochondrial DNA-depleted microbial sequencing reads; and (d) purifying the mitochondrial DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads. In some embodiments, purification includes in silico purification. In some embodiments, the genomic signatures include microbial phylogenetic marker genes or marker gene fragments thereof. In some embodiments, the microbial phylogenetic marker genes include bacterial marker genes or marker gene fragments thereof. In some embodiments, the microbial phylogenetic marker genes include fungal marker genes or marker gene fragments thereof. In some embodiments, the bacterial marker genes include ribosomal RNA gene 5S, ribosomal RNA gene 16S, ribosomal RNA gene 23S, bacterial housekeeping genes dnaG, frr, infC, nusA, pgk, pyrG, rplA, rplB, rplC, rplD, rplE, rplF, rplK, rplL, rplM, rplN, rplP, rplS, rplT, rpmA, rpoB, rpsB, rpsC, rpsE, rpsI, rpsJ, rpsK, rpsM, rpsS, smpB, tsf, or any combination thereof. In some embodiments, the fungal marker genes include ribosomal RNA gene 18S, ribosomal RNA gene 5.8S, ribosomal RNA gene 28S, and one or more of the internally transcribed spacer regions 1 and 2. In some embodiments, the microbial phylogenetic marker genes include marker genes of bacteria, fungi, or any combination thereof. In some embodiments, amplification includes performing polymerase chain reaction or a derivative thereof. In some embodiments, the derivative includes inverse PCR, anchored PCR, primer-directed rolling circle amplification, or any combination thereof.

[0052] In some embodiments, step (c) of the method of generating a microbial metagenomic signature set for diagnosing cancer comprises enriching the microbial nucleic acid molecules. In some embodiments, the enrichment comprises (a) combining the purified nucleosome-depleted microbial nucleic acid molecules with a hybridization probe, wherein the hybridization probe comprises nucleic acid sequence complementarity to a microbial genomic signature; (b) incubating the hybridization probe and the nucleosome-depleted microbial nucleic acid molecules under conditions that promote nucleic acid base pairing between the target nucleic acid signature and the hybridization probe; (c) separating unbound hybridization probe and hybridized probes bound to the microbial nucleic acid molecules; and (d) washing the hybridized probes bound to the microbial nucleic acid molecules, thereby generating enriched microbial nucleic acid molecules. In some embodiments, the washing is configured to remove non-specifically associated nucleic acid molecules and other reaction components.

[0053] In some embodiments, enrichment comprises: (a) combining a purified nucleosome-depleted microbial nucleic acid molecule with a recombinant CXXC domain protein to form a protein-DNA binding reaction; (b) incubating the protein-DNA binding reaction under conditions that promote interaction between the recombinant CXXC domain protein and the unmethylated CpG motif of the nucleosome-depleted microbial nucleic acid molecule; (c) separating unbound recombinant CXXC domain protein and recombinant CXXC domain protein bound to the unmethylated CpG motif from the remainder of the protein-DNA binding reaction; (d) washing the recombinant CXXC domain protein bound to the unmethylated CpG nucleic acid fragment, thereby generating an enriched nucleic acid molecule for amplification. In some embodiments, the washing is configured to remove non-specifically associated nucleic acid molecules and the remainder of the protein-DNA binding reaction components. In some embodiments, amplification comprises performing a polymerase chain reaction or a derivative thereof. In some embodiments, the derivative comprises inverse PCR, anchored PCR, primer-directed rolling circle amplification, or any combination thereof. In some embodiments, the mammalian nucleic acid molecule and the microbial nucleic acid molecule are derived from the liquid biological sample of the plurality of subjects. In some embodiments, the plurality of subjects comprises subjects that are human, non-human mammals, or any combination thereof. In some embodiments, the mammalian nucleic acid molecule comprises nucleic acid molecules that are DNA, RNA, cell-free DNA, cell-free RNA, exosomal DNA, exosomal RNA, or any combination thereof, and the microbial nucleic acid molecule comprises nucleic acid molecules that are microbial cell-free RNA, microbial cell-free DNA, microbial RNA, microbial DNA, or any combination thereof.In some embodiments, the cancer includes acute myeloid leukemia, adrenocortical carcinoma, bladder urothelial carcinoma, brain low-grade glioma, breast invasive cancer, cervical squamous cell carcinoma and endocervical adenocarcinoma, cholangiocarcinoma, colorectal adenocarcinoma, esophageal cancer, glioblastoma multiforme, head and neck squamous cell carcinoma, kidney chromophobe cell, kidney clear cell renal carcinoma, kidney papillary renal carcinoma, liver hepatocellular carcinoma, lung adenocarcinoma, lung squamous cell carcinoma, lymphoma diffuse large B-cell lymphoma, mesothelioma, ovarian serous cystadenocarcinoma, pancreatic adenocarcinoma, pheochromocytoma and paraganglioma, prostate adenocarcinoma, rectal adenocarcinoma, sarcoma, skin melanoma, gastric adenocarcinoma, testicular germ cell tumor, thymoma, thyroid cancer, uterine carcinosarcoma, endometrial carcinoma of the uterine body, choroidal melanoma, or any combination thereof. In some embodiments, the cancer includes stage I, II, or III cancer. In some embodiments, the method of generating a microbial metagenomic signature set for diagnosing cancer includes generating a trained prediction model, and the trained prediction model is trained with the microbial metagenomic signature set of the one or more subjects and the health state. In some embodiments, the trained prediction model includes a machine learning model, one or more machine learning models, an ensemble of machine learning models, or any combination thereof. In some embodiments, the trained prediction model includes a regularized machine learning model. In some embodiments, the machine learning model includes a machine learning classifier. In some embodiments, the machine learning model includes a gradient boosting machine, neural network, support vector machine, k-means, classification tree, random forest, regression, or any combination thereof machine learning model. In some embodiments, the health state of the subject or subjects includes a known non-neoplastic disease, cancer, or any combination thereof of the subject.

[0054] Aspects of the present disclosure include a method for diagnosing a target cancer, the method comprising: (a) providing a liquid biological sample of the subject, wherein the liquid biological sample comprises mammalian nucleic acid molecules and microbial nucleic acid molecules; (b) removing the mammalian nucleic acid molecules from the liquid biological sample using an affinity capture reagent; (c) sequencing a plurality of microbial nucleic acid molecules of the liquid biological sample to generate microbial sequencing reads; (d) generating a metagenomic feature abundance of the microbial sequencing reads; and (e) outputting a diagnosis of the cancer of the subject as a result of providing at least the microbial metagenomic feature abundance as an input to a trained prediction model. In some embodiments, the microbial metagenomic feature set includes microbial taxonomic abundances. In some embodiments, the microbial metagenomic feature set includes computationally inferred microbial biochemical pathways and their associated abundances. In some embodiments, the microbial metagenomic feature set includes microbial phylogenetic marker genes or marker gene fragments thereof. In some embodiments, the liquid biological sample includes plasma, serum, whole blood, urine, cerebrospinal fluid, saliva, sweat, tears, exhaled condensate, or any combination, dilution, or processed fraction thereof.

[0055] In some embodiments, step (b) of the method for diagnosing a target cancer comprises: (a) contacting the liquid biological sample with a solid support comprising an immobilized anti-nucleosome antibody, wherein the anti-nucleosome antibody is configured to form an antibody-nucleosome interaction complex; (b) separating the solid support from the liquid biological sample to concentrate the antibody-nucleosome interaction complex; and (c) purifying the remaining nucleosome-depleted microbial nucleic acid molecules. In some embodiments, the anti-nucleosome antibody recognizes an epitope comprising DNA and one or more histone proteins. In some embodiments, the solid support includes magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, pH-sensitive polymers, or any combination thereof.

[0056] In some embodiments, step (b) of the method for diagnosing a target cancer comprises: (a) contacting the liquid biological sample with an anti-nucleosome antibody to form an antibody-nucleosome interaction complex; (b) contacting the antibody-nucleosome interaction complex with a solid support configured to bind to the antibody-nucleosome interaction complex; (c) separating the solid support from the liquid biological sample to concentrate the antibody-nucleosome interaction complex; and (d) purifying the remaining nucleosome-depleted microbial nucleic acids. In some embodiments, the anti-nucleosome antibody comprises an epitope tag. In some embodiments, the epitope tag comprises an N-terminal or C-terminal 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), Fc fusion, biotin, or any combination thereof. In some embodiments, the solid support comprises magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, pH-sensitive polymer, or any combination thereof. In some embodiments, the solid support comprises an affinity agent immobilized by covalent bond. In some embodiments, the affinity agent immobilized by covalent bond comprises streptavidin, 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), an antibody specific for biotin, or any combination thereof. In some embodiments, the affinity agent immobilized by covalent bond comprises an anti-species antibody.

[0057] In some embodiments, step (c) of the method for diagnosing a target cancer comprises: (a) generating a single-stranded DNA library from the microbial nucleic acid molecules; (b) performing shotgun metagenomic sequencing analysis of the single-stranded DNA library to generate sequencing reads; (c) filtering the sequencing reads to generate mammalian DNA-depleted microbial sequencing reads; and (d) purifying the mammalian DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads. In some embodiments, the purification comprises in silico purification of the mammalian DNA-depleted microbial sequencing reads. In some embodiments, the filtering comprises computationally mapping the sequencing reads to a human reference genome database.

[0058] In some embodiments, step (c) of the method for diagnosing a target cancer comprises: (a) amplifying genomic features of the microbial nucleic acid molecule, thereby generating amplified genomic features; (b) sequencing the amplified genomic features to generate sequencing reads; (c) filtering the sequencing reads to generate mitochondrial DNA-depleted microbial sequencing reads; and (d) purifying the mitochondrial DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads. In some embodiments, purification includes in silico purification of the mitochondrial DNA-depleted microbial sequencing reads. In some embodiments, the genomic features include a microbial phylogenetic marker gene or a marker gene fragment thereof. In some embodiments, the microbial phylogenetic marker gene includes a bacterial marker gene or a marker gene fragment thereof. In some embodiments, the microbial phylogenetic marker gene includes a fungal marker gene or a marker gene fragment thereof. In some embodiments, the bacterial marker gene includes ribosomal RNA gene 5S, ribosomal RNA gene 16S, ribosomal RNA gene 23S, bacterial housekeeping gene dnaG, frr, infC, nusA, pgk, pyrG, rplA, rplB, rplC, rplD, rplE, rplF, rplK, rplL, rplM, rplN, rplP, rplS, rplT, rpmA, rpoB, rpsB, rpsC, rpsE, rpsI, rpsJ, rpsK, rpsM, rpsS, smpB, tsf, or any combination thereof. In some embodiments, the fungal marker gene includes ribosomal RNA gene 18S, ribosomal RNA gene 5.8S, ribosomal RNA gene 28S, and one or more of internal transcribed spacer regions 1 and 2. In some embodiments, the microbial phylogenetic marker gene includes marker genes of bacteria, fungi, or any combination thereof. In some embodiments, amplification includes performing polymerase chain reaction or a derivative thereof. In some embodiments, the derivative includes inverse PCR, anchor PCR, primer-directed rolling circle amplification, or any combination thereof.

[0059] In some embodiments, step (c) of the method for diagnosing a target cancer comprises enriching the microbial nucleic acid molecule. In some embodiments, the enrichment of the microbial nucleic acid molecule comprises: (a) combining a purified nucleosome-depleted microbial nucleic acid molecule with a hybridization probe, wherein the hybridization probe comprises nucleic acid sequence complementarity to a microbial genomic nucleic acid feature; (b) incubating the hybridization probe and the nucleosome-depleted microbial nucleic acid molecule under conditions that promote base pairing between the microbial genomic nucleic acid feature and the hybridization probe; (c) separating unbound hybridization probe and hybridized probe bound to the nucleosome-depleted microbial nucleic acid molecule; and (d) washing the hybridized probe bound to the nucleosome-depleted microbial nucleic acid molecule, thereby generating one or more enriched microbial nucleic acid molecules. In some embodiments, the washing is configured to remove non-specifically associated nucleic acid molecules and other reaction components.

[0060] In some embodiments, the enrichment of the microbial nucleic acid molecules comprises: (a) combining the purified nucleosome-depleted microbial nucleic acid molecules with a recombinant CXXC domain protein to form a protein-DNA binding reaction; (b) incubating the protein-DNA binding reaction under conditions that promote the interaction between the recombinant CXXC domain protein and the non-methylated CpG motifs of the nucleosome-depleted microbial nucleic acid molecules; (c) separating the unbound recombinant CXXC domain protein and the recombinant CXXC domain protein bound to the non-methylated CpG nucleic acid fragments from the rest of the protein-DNA binding reaction components; (d) washing the recombinant CXXC domain protein bound to the non-methylated CpG nucleic acid fragments, thereby generating enriched nucleic acid molecules for amplification. In some embodiments, the washing is configured to remove non-specifically associated nucleic acid molecules and the rest of the protein-DNA binding reaction components. In some embodiments, the amplification comprises performing a polymerase chain reaction or a derivative thereof. In some embodiments, the derivative comprises inverse PCR, anchored PCR, primer-directed rolling circle amplification, or any combination thereof. In some embodiments, the mammalian nucleic acid molecules and the microbial nucleic acid molecules are derived from a liquid biological sample of the subject. In some embodiments, the subject comprises a human, non-human mammal, or any combination thereof. In some embodiments, the mammalian nucleic acid molecules comprise nucleic acid molecules of DNA, RNA, cell-free RNA, cell-free DNA, exosomal DNA, exosomal RNA, or any combination thereof, and the microbial nucleic acid molecules comprise nucleic acid molecules of microbial cell-free DNA, microbial cell-free RNA, microbial DNA, microbial RNA, or any combination thereof.In some embodiments, the cancer includes acute myeloid leukemia, adrenocortical carcinoma, bladder urothelial carcinoma, low-grade glioma of the brain, invasive breast cancer, squamous cell carcinoma and adenocarcinoma of the cervix, cholangiocarcinoma, colorectal adenocarcinoma, esophageal cancer, glioblastoma multiforme, squamous cell carcinoma of the head and neck, renal chromophobe cell, renal clear cell carcinoma of the kidney, renal papillary cell carcinoma of the kidney, hepatocellular carcinoma of the liver, lung adenocarcinoma, lung squamous cell carcinoma, lymphoma diffuse large B-cell lymphoma, mesothelioma, serous cystadenocarcinoma of the ovary, pancreatic adenocarcinoma, pheochromocytoma and paraganglioma, prostate adenocarcinoma, rectal adenocarcinoma, sarcoma, cutaneous melanoma of the skin, gastric adenocarcinoma, testicular germ cell tumor, thymoma, thyroid cancer, uterine carcinosarcoma, endometrial cancer of the uterine body, uveal melanoma, or any combination thereof. In some embodiments, the cancer includes stage I, II, or III cancer. In some embodiments, the trained prediction model is trained with a microbial metagenomic feature set of one or more subjects and the corresponding health state. In some embodiments, the trained prediction model includes a machine learning model, one or more machine learning models, an ensemble of machine learning models, or any combination thereof. In some embodiments, the trained prediction model includes a regularized machine learning model. In some embodiments, the machine learning model includes a machine learning classifier. In some embodiments, the machine learning model includes a gradient boosting machine, a neural network, a support vector machine, k-means, a classification tree, a random forest, regression, or any combination thereof of machine learning models. In some embodiments, the subject is suspected of having cancer or a disease. In some embodiments, the result of imaging of the subject indicates the potential presence of cancer. In some embodiments, the health state of the subject or subjects includes a known non-neoplastic disease, cancer, or any combination thereof of the subject.

[0061] Aspects of the present disclosure include a system for diagnosing a target cancer, the system comprising: (a) a processor; and (b) non-transitory computer-readable storage medium including software configured to cause the processor to: (i) receive mammalian nucleosome-depleted nucleic acid molecule sequencing reads of a subject, the mammalian nucleosome-depleted nucleic acid molecule sequencing reads including metagenomic features of microbial nucleic acid molecules; and (ii) output a diagnosis of the cancer of the subject as a result of providing at least the metagenomic features as an input to a trained prediction model. In some embodiments, the metagenomic features include microbial taxonomic abundances. In some embodiments, the metagenomic features include computationally inferred microbial biochemical pathways and their associated abundances. In some embodiments, the metagenomic features include microbial phylogenetic marker genes or marker gene fragments thereof. In some embodiments, the liquid biological sample includes plasma, serum, whole blood, urine, cerebrospinal fluid, saliva, sweat, tears, exhaled condensate, or any combination, dilution, or processed fraction thereof. In some embodiments, the mammalian nucleosome-depleted nucleic acid molecule sequencing reads are generated by: (a) contacting the liquid biological sample with a solid support to form an antibody-nucleosome interaction complex, the solid support including a surface coupled with an anti-nucleosome antibody on the surface; (b) separating the solid support from the liquid biological sample to concentrate the antibody-nucleosome interaction complex; (c) purifying the remaining nucleosome-depleted microbial nucleic acid molecules; and (d) sequencing the purified nucleosome-depleted microbial nucleic acid molecules. In some embodiments, the anti-nucleosome antibody is configured to recognize an epitope including DNA and histone proteins. In some embodiments, the solid support includes magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, pH-sensitive polymers, or any combination thereof.

[0062] In some embodiments, the sequencing reads of mammalian nucleosome-depleted nucleic acid molecules are generated by: (a) contacting the liquid biological sample with an anti-nucleosome antibody to form an antibody-nucleosome interaction complex; (b) contacting the antibody-nucleosome interaction complex with a solid support; (c) separating the solid support from the liquid biological sample to concentrate the antibody-nucleosome interaction complex; (d) purifying the remaining nucleosome-depleted microbial nucleic acid molecules; and (e) sequencing the one or more purified nucleosome-depleted microbial nucleic acid molecules. In some embodiments, the anti-nucleosome antibody comprises an epitope tag. In some embodiments, the epitope tag comprises an N-terminal or C-terminal 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), an Fc fusion, biotin, or any combination thereof. In some embodiments, the solid support comprises magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, a pH-sensitive polymer, or any combination thereof. In some embodiments, the solid support comprises a covalently immobilized affinity agent. In some embodiments, the covalently immobilized affinity agent comprises streptavidin, a 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), an antibody specific for biotin, or any combination thereof. In some embodiments, the covalently immobilized affinity agent comprises an anti-species antibody.

[0063] In some embodiments, sequencing reads of mammalian nucleosome-depleted nucleic acid molecules are generated by: (a) generating a single-stranded DNA library from the microbial nucleic acid molecules; (b) performing shotgun metagenomic sequencing analysis of the single-stranded DNA library to generate sequencing reads; (c) filtering the sequencing reads to generate mammalian DNA-depleted microbial sequencing reads; and (d) purifying the mammalian DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads. In some embodiments, purification includes in silico purification of the mammalian DNA-depleted microbial sequencing reads. In some embodiments, filtering includes computationally mapping the sequencing reads to a human reference genome database.

[0064] In some embodiments, mammalian nucleosome-depleted nucleic acid molecule sequencing reads are generated by (a) amplifying genomic features of the one or more microbial nucleic acid molecules, thereby generating amplified genomic features; (b) sequencing the amplified genomic features to generate sequencing reads; (c) filtering the sequencing reads to produce mitochondrial DNA-depleted microbial sequencing reads; and (d) purifying the mitochondrial DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads. In some embodiments, purification includes in silico purification of the mitochondrial DNA-depleted microbial sequencing reads. In some embodiments, the genomic features include microbial phylogenetic marker genes or marker gene fragments thereof. In some embodiments, the microbial phylogenetic marker genes include bacterial marker genes or marker gene fragments thereof. In some embodiments, the microbial phylogenetic marker genes include fungal marker genes or marker gene fragments thereof. In some embodiments, the bacterial marker genes include ribosomal RNA gene 5S, ribosomal RNA gene 16S, ribosomal RNA gene 23S, bacterial housekeeping genes dnaG, frr, infC, nusA, pgk, pyrG, rplA, rplB, rplC, rplD, rplE, rplF, rplK, rplL, rplM, rplN, rplP, rplS, rplT, rpmA, rpoB, rpsB, rpsC, rpsE, rpsI, rpsJ, rpsK, rpsM, rpsS, smpB, tsf, or any combination thereof. In some embodiments, the fungal marker genes include ribosomal RNA gene 18S, ribosomal RNA gene 5.8S, ribosomal RNA gene 28S, and one or more of internal transcribed spacer regions 1 and 2. In some embodiments, the microbial phylogenetic marker genes include marker genes of bacteria, fungi, or any combination thereof. In some embodiments, amplification includes performing polymerase chain reaction (PCR) or a derivative thereof. In some embodiments, the derivative includes inverse PCR, anchored PCR, primer-directed rolling circle amplification, or any combination thereof.In some embodiments, the microbial nucleic acid molecules are enriched from the mammalian nucleosome-depleted nucleic acid molecules.

[0065] In some embodiments, the enrichment of the microbial nucleic acid molecules comprises: (a) combining a purified nucleosome-depleted microbial nucleic acid molecule with a hybridization probe, wherein the hybridization probe comprises nucleic acid sequence complementarity to a microbial genomic nucleic acid feature; (b) incubating the hybridization probe and the nucleosome-depleted microbial nucleic acid molecule under conditions that promote nucleic acid base pairing between the microbial genomic nucleic acid feature and the hybridization probe; (c) separating unbound hybridization probe and hybridized probe bound to the nucleosome-depleted microbial nucleic acid molecule; and (d) washing the hybridized probe bound to the nucleosome-depleted microbial nucleic acid molecule, thereby generating enriched microbial nucleic acid molecules. In some embodiments, the washing is configured to remove non-specifically associated nucleic acid molecules and other reaction components.

[0066] In some embodiments, the enrichment of the microbial nucleic acid molecules comprises: (a) combining the purified nucleosome-depleted microbial nucleic acid molecules with a recombinant CXXC domain protein to form a protein-DNA binding reaction; (b) incubating the protein-DNA binding reaction under conditions that promote the interaction between the recombinant CXXC domain protein and the unmethylated CpG motifs of the nucleosome-depleted microbial nucleic acid molecules; (c) separating the unbound recombinant CXXC domain protein and the recombinant CXXC domain protein bound to the unmethylated CpG nucleic acid fragments from the remaining protein-DNA binding reaction components; and (d) washing the recombinant CXXC domain protein bound to the unmethylated CpG nucleic acid fragments, thereby generating enriched nucleic acid molecules for amplification. In some embodiments, the washing is configured to remove nonspecifically associated nucleic acid molecules and the remaining protein-DNA binding reaction components. In some embodiments, the amplification comprises performing a polymerase chain reaction (PCR) or a derivative thereof. In some embodiments, the derivative comprises inverse PCR, anchor PCR, primer-directed rolling circle amplification, or any combination thereof. In some embodiments, the mammalian nucleic acid molecules and the microbial nucleic acid molecules are derived from a liquid biological sample of the subject. In some embodiments, the subject comprises a human, non-human mammal, or any combination thereof. In some embodiments, the mammalian nucleic acid molecules comprise DNA, RNA, cell-free RNA, cell-free DNA, exosomal DNA, exosomal RNA, or any combination thereof, and the microbial nucleic acid molecules comprise microbial cell-free DNA, microbial cell-free RNA, microbial DNA, microbial RNA, or any combination thereof.In some embodiments, the cancer includes acute myeloid leukemia, adrenocortical carcinoma, bladder urothelial carcinoma, brain low-grade glioma, breast invasive cancer, cervical squamous cell carcinoma and endocervical adenocarcinoma, cholangiocarcinoma, colon adenocarcinoma, esophageal cancer, glioblastoma multiforme, head and neck squamous cell carcinoma, kidney chromophobe cell, kidney clear cell renal carcinoma, kidney papillary renal cell carcinoma, liver hepatocellular carcinoma, lung adenocarcinoma, lung squamous cell carcinoma, lymphoma diffuse large B-cell lymphoma, mesothelioma, ovarian serous cystadenocarcinoma, pancreatic adenocarcinoma, pheochromocytoma and paraganglioma, prostate adenocarcinoma, rectal adenocarcinoma, sarcoma, skin melanoma, stomach adenocarcinoma, testicular germ cell tumor, thymoma, thyroid cancer, uterine carcinosarcoma, endometrial carcinoma of the uterine corpus, choroidal melanoma, or any combination thereof. In some embodiments, the cancer includes stage I, II, or III cancer. In some embodiments, the trained prediction model is trained with metagenomic features and corresponding health states of a plurality of subjects. In some embodiments, the trained prediction model includes a machine learning model, one or more machine learning models, an ensemble of machine learning models, or any combination thereof. In some embodiments, the trained prediction model includes a regularized machine learning model. In some embodiments, the machine learning model includes a machine learning classifier. In some embodiments, the machine learning model includes a gradient boosting machine, neural network, support vector machine, k-means, classification tree, random forest, regression, or any combination thereof of machine learning models. In some embodiments, the subject is suspected of having cancer or a disease. In some embodiments, the results of imaging of the subject indicate the potential presence of cancer.

[0067] Aspects of the present disclosure are methods for generating metagenomic features of a sample of cell-free microbial nucleic acid molecules for diagnosing non-neoplastic diseases, comprising: (a) contacting a sample of the cell-free nucleic acid molecules with a probe, the probe comprising a binding site configured to bind to a human nucleic acid molecule complexed with a protein; (b) removing the probe bound to the human nucleic acid molecule complexed with the protein, thereby producing an enriched cell-free microbial nucleic acid molecule; and (c) generating metagenomic features of the enriched cell-free microbial nucleic acid molecule configured to diagnose non-neoplastic diseases. In some embodiments, the protein comprises one or more histone proteins, one or more regulatory proteins, or any combination thereof. In some embodiments, the sample comprises plasma, serum, whole blood, urine, cerebrospinal fluid, saliva, sweat, tears, breath condensate, or any combination, dilution, or processed fraction thereof. In some cases, the probe comprises one or more antibodies. In some embodiments, the removal comprises incubating the antibody bound to the human nucleic acid molecule complexed with the protein with a solid support, the solid support comprising a capture reagent configured to bind to the antibody.

[0068] In some embodiments, step (c) of the method of generating a metagenomic profile of a sample of cell-free microbial nucleic acids for diagnosing a non-neoplastic disease comprises contacting the enriched cell-free microbial nucleic acids with a second set of probes, wherein the second set of probes is configured to bind to microbial marker genes. In some embodiments, the microbial marker genes include ribosomal RNA gene 5S, ribosomal RNA gene 16S, ribosomal RNA gene 23S, bacterial housekeeping gene dnaG, frr, infC, nusA, pgk, pyrG, rplA, rplB, rplC, rplD, rplE, rplF, rplK, rplL, rplM, rplN, rplP, rplS, rplT, rpmA, rpoB, rpsB, rpsC, rpsE, rpsI, rpsJ, rpsK, rpsM, rpsS, smpB, tsf, or any combination thereof. In some embodiments, the microbial marker genes are sequenced to determine the abundance of the taxonomy, function, or any combination thereof of the microorganisms. In some embodiments, the sample includes a liquid biological sample. In some embodiments, the sample is derived from a subject. In some embodiments, the subject is a human or non-human mammal. In some embodiments, the protein includes histone proteins associated with nucleic acid molecules. In some embodiments, the human nucleic acid molecules include DNA, RNA, cell-free RNA, cell-free DNA, exosomal RNA, exosomal DNA, or any combination thereof. In some embodiments, the cell-free microbial nucleic acids include cell-free microbial DNA, cell-free microbial RNA, microbial RNA, microbial DNA, or any combination thereof. In some embodiments, the removal comprises immunoprecipitating the probes bound to the human nucleic acid molecules. In some embodiments, the method of generating a metagenomic profile of a sample of cell-free microbial nucleic acids for diagnosing a non-neoplastic disease further comprises (d) preparing a single-stranded library from the cell-free microbial nucleic acids of the sample. In some embodiments, the first set of probes is coupled to a solid support.In some embodiments, the solid support comprises beads, magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, pH-sensitive polymers, or any combination thereof. In some embodiments, the sample comprises human nucleic acid molecules, microbial nucleic acid molecules, or any combination thereof. In some embodiments, the non-neoplastic disease comprises a benign tumor of the integumentary, skeletal, muscular, nervous, endocrine, cardiovascular, lymphatic, digestive, respiratory, urinary, reproductive, or any combination of those systems. In some embodiments, the health condition of the subject or subjects comprises a known non-neoplastic disease of the subject, cancer, or any combination thereof.

[0069] Aspects of the present disclosure include a method of generating a microbial metagenomic signature set for diagnosing a non-neoplastic disease, the method comprising: (a) providing a plurality of subject health conditions and a plurality of subject liquid biological samples, wherein the liquid biological samples comprise mammalian nucleic acid molecules and microbial nucleic acid molecules; (b) removing the mammalian nucleic acid molecules from the liquid biological samples using an affinity capture reagent; (c) sequencing the microbial nucleic acid molecules to generate microbial sequencing reads; and (d) generating the microbial metagenomic signature set for diagnosing a non-neoplastic disease by combining the metagenomic signature abundance of the microbial sequencing reads with the plurality of subject health conditions. In some embodiments, the metagenomic signature set comprises microbial taxonomic abundances. In some embodiments, the metagenomic signature set comprises computationally inferred microbial biochemical pathways and associated abundances of the microbial biochemical pathways. In some embodiments, the metagenomic signature set comprises microbial phylogenetic marker genes or marker gene fragments thereof. In some embodiments, the liquid biological sample comprises plasma, serum, whole blood, urine, cerebrospinal fluid, saliva, sweat, tears, breath condensate, or any combination, dilution, or processed fraction thereof.

[0070] In some embodiments, step (b) of the method of generating a microbial metagenomic signature set for diagnosing a non-neoplastic disease comprises: (a) contacting the liquid biological sample with a solid support comprising an immobilized anti-nucleosome antibody to form an antibody-nucleosome interaction complex; (b) separating the solid support from the liquid biological sample to concentrate the antibody-nucleosome interaction complex; and (c) purifying the remaining one or more nucleosome-depleted microbial nucleic acid molecules. In some embodiments, the anti-nucleosome antibody is configured to bind to an epitope comprising DNA and one or more histone proteins. In some embodiments, the solid support comprises magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, pH-sensitive polymers, or any combination thereof.

[0071] In some embodiments, step (b) of the method for generating a microbial metagenomic signature set for diagnosing a non-neoplastic disease comprises: (a) contacting the liquid biological sample with one or more anti-nucleosome antibodies to form an antibody-nucleosome interaction complex; (b) contacting the antibody-nucleosome interaction complex with a solid support; (c) separating the solid support from the liquid biological sample to concentrate the antibody-nucleosome interaction complex; and (d) purifying the remaining nucleosome-depleted microbial nucleic acid molecules. In some embodiments, the anti-nucleosome antibody comprises a plurality of epitope tags. In some embodiments, the plurality of epitope tags comprises an N-terminal or C-terminal 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), an Fc fusion, biotin, or any combination thereof. In some embodiments, the solid support comprises magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, a pH-sensitive polymer, or any combination thereof. In some embodiments, the solid support comprises an affinity agent immobilized by covalent bonding. In some embodiments, the affinity reagent comprises streptavidin, a 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), an antibody specific for biotin, or any combination thereof. In some embodiments, the affinity agent comprises an anti-species antibody.

[0072] In some embodiments, step (c) of the method for generating a microbial metagenomic feature set for diagnosing a non-tumorous disease includes: (a) generating a single-stranded DNA library from the microbial nucleic acid molecules; (b) performing shotgun metagenomic sequencing analysis on the single-stranded DNA library to generate sequencing reads; (c) filtering the sequencing reads to generate mammalian DNA-depleted microbial sequencing reads; and (d) purifying the mammalian DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads. In some embodiments, the purification includes in silico purification. In some embodiments, the filtering includes computationally mapping the sequencing reads to a human reference genome database.

[0073] In some embodiments, step (c) of the method for generating a microbial metagenomic feature set for diagnosing a non-neoplastic disease comprises: (a) amplifying genomic features of the microbial nucleic acid molecules, thereby generating amplified genomic features; (b) sequencing the amplified genomic features to generate sequencing reads; filtering the sequencing reads to produce mitochondrial DNA-depleted microbial sequencing reads; and (c) purifying the mitochondrial DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads. In some embodiments, purification includes in silico purification. In some embodiments, the genomic features include microbial phylogenetic marker genes or marker gene fragments thereof. In some embodiments, the microbial phylogenetic marker genes include bacterial marker genes or marker gene fragments thereof. In some embodiments, the microbial phylogenetic marker genes include fungal marker genes or marker gene fragments thereof. In some embodiments, the bacterial marker genes include ribosomal RNA gene 5S, ribosomal RNA gene 16S, ribosomal RNA gene 23S, bacterial housekeeping genes dnaG, frr, infC, nusA, pgk, pyrG, rplA, rplB, rplC, rplD, rplE, rplF, rplK, rplL, rplM, rplN, rplP, rplS, rplT, rpmA, rpoB, rpsB, rpsC, rpsE, rpsI, rpsJ, rpsK, rpsM, rpsS, smpB, tsf, or any combination thereof. In some embodiments, the fungal marker genes include ribosomal RNA gene 18S, ribosomal RNA gene 5.8S, ribosomal RNA gene 28S, and one or more of the internal transcribed spacer regions 1 and 2. In some embodiments, the microbial phylogenetic marker genes include marker genes of bacteria, fungi, or any combination thereof. In some embodiments, amplification includes performing polymerase chain reaction or a derivative thereof. In some embodiments, the derivative includes inverse PCR, anchored PCR, primer-directed rolling circle amplification, or any combination thereof.

[0074] In some embodiments, step (c) of the method of generating a microbial metagenomic signature set for diagnosing a non - neoplastic disease comprises enriching the microbial nucleic acid molecules. In some embodiments, the enrichment comprises: (a) combining the purified nucleosome - depleted microbial nucleic acid molecules with a hybridization probe, wherein the hybridization probe comprises nucleic acid sequence complementarity to a microbial genomic signature; (b) incubating the hybridization probe and the nucleosome - depleted microbial nucleic acid molecules under conditions that promote nucleic acid base pairing between the target nucleic acid signature and the hybridization probe; (c) separating the unbound hybridization probe and the hybridized probe bound to the microbial nucleic acid molecules; and (d) washing the hybridized probe bound to the microbial nucleic acid molecules, thereby generating enriched microbial nucleic acid molecules. In some embodiments, the washing is configured to remove non - specifically associated nucleic acid molecules and other reaction components.

[0075] In some embodiments, enrichment comprises: (a) combining a purified nucleosome-depleted microbial nucleic acid molecule with a recombinant CXXC domain protein to form a protein-DNA binding reaction; (b) incubating the protein-DNA binding reaction under conditions that promote the interaction between the recombinant CXXC domain protein and the unmethylated CpG motif of the nucleosome-depleted microbial nucleic acid molecule; (c) separating unbound recombinant CXXC domain protein and recombinant CXXC domain protein bound to the unmethylated CpG motif from the remainder of the protein-DNA binding reaction; and (d) washing the recombinant CXXC domain protein bound to the unmethylated CpG nucleic acid fragment, thereby generating an enriched nucleic acid molecule for amplification. In some embodiments, the washing is configured to remove non-specifically associated nucleic acid molecules and the remainder of the protein-DNA binding reaction components. In some embodiments, amplification comprises performing polymerase chain reaction or a derivative thereof. In some embodiments, the derivative comprises inverse PCR, anchored PCR, primer-directed rolling circle amplification, or any combination thereof. In some embodiments, the mammalian nucleic acid molecule and the microbial nucleic acid molecule are derived from the liquid biological samples of the plurality of subjects. In some embodiments, the plurality of subjects comprises subjects that are human, non-human mammalian, or any combination thereof. In some embodiments, the mammalian nucleic acid molecule comprises nucleic acid molecules that are DNA, RNA, cell-free DNA, cell-free RNA, exosomal DNA, exosomal RNA, or any combination thereof, and the microbial nucleic acid molecule comprises nucleic acid molecules that are microbial cell-free RNA, microbial cell-free DNA, microbial RNA, microbial DNA, or any combination thereof.

[0076] In some embodiments, a method of generating a microbial metagenomic feature set for diagnosing a non-neoplastic disease includes generating a trained prediction model, the trained prediction model being trained on the microbial metagenomic feature set of the one or more subjects and the health state. In some embodiments, the trained prediction model includes a machine learning model, one or more machine learning models, an ensemble of machine learning models, or any combination thereof. In some embodiments, the trained prediction model includes a regularized machine learning model. In some embodiments, the machine learning model includes a machine learning classifier. In some embodiments, the machine learning model includes a gradient boosting machine, a neural network, a support vector machine, k-means, a classification tree, a random forest, regression, or any combination thereof of machine learning models. In some embodiments, the non-neoplastic disease includes a benign tumor of the integumentary, skeletal, muscular, nervous, endocrine, cardiovascular, lymphatic, digestive, respiratory, urinary, reproductive, or any combination of those systems.

[0077] In some embodiments, the health state of the subject or subjects includes a known non-neoplastic disease of the subject, cancer, or any combination thereof.

[0078] The novel features of the invention are set forth in detail in the appended claims. A better understanding of the features and advantages of the present disclosure will be obtained by reference to the following detailed description, which describes exemplary embodiments in which the principles of the present disclosure are used, and the accompanying drawings.

Brief Description of the Drawings

[0079]

Fig. 1

Fig. 2A

Fig. 2B

Fig. 3A

Fig. 3B

Fig. 4A

Fig. 4B

Fig. 5

Fig. 6A

Fig. 6B

Fig. 6C

Fig. 7

DETAILED DESCRIPTION OF THE INVENTION

[0080] It has been shown that the abundance of unique microorganisms (e.g., bacteria, viruses, and / or fungi) can function as a proxy for the diagnosis or screening of a disease, such as cancer, cancer stage, the organ of origin of cancer, cancer subtype, or any combination thereof. The microbial abundance(s) associated with and / or correlated with such cancer can be determined by detecting the abundance of non-human nucleic acid molecules of cell-free microbial nucleic acids in a biological sample of a subject (e.g., a solid and / or liquid biological sample), for example, through low-invasive screening and / or diagnosis. Detecting the microbial abundance from cell-free microbial nucleic acid molecules in a biological sample of a subject faces many problems due to the essentially small amount of cell-free microbial nucleic acid molecules present in a large amount of host human nucleic acid molecules. Therefore, there is a need for a method and / or system configured to deplete the host human nucleic acid molecules (e.g., human DNA and / or RNA) of a biological sample in order to enrich the relatively small amount of cell-free microbial nucleic acid molecules present in the biological sample. The abundance of the enriched cell-free microbial nucleic acid molecules can be used to generate one or more features that correlate and / or are related to the health status (e.g., cancerous, non-cancerous disease, or healthy) of the subject from which the cell-free microbial nucleic acid molecules were obtained and / or collected as described elsewhere in this specification. In some cases, a prediction model (e.g., an artificial intelligence model, a machine learning model, etc.) can be trained with one or more features of the microbial nucleic acid molecule abundance, and the trained prediction model can be used to predict, monitor, and / or diagnose the health status of the microbial nucleic acid molecule abundance(s) of a biological sample of a subject not used to train the prediction model.

[0081] Methods and systems are provided herein that can be configured to enrich microbial nucleic acid molecules and / or deplete human nucleic acid molecules of a biological sample of a subject for use in disease screening and / or diagnostic assays described elsewhere herein. Enrichment of microbial nucleic acid molecules and / or depletion of human nucleic acid molecules can increase the accuracy, specificity, and / or sensitivity of the diagnosis, screening, and / or prediction results provided by the prediction models described elsewhere herein by at least about 1%, at least about 2%, at least about 3%, at least about 4%, at least about 5%, at least about 6%, at least about 7%, at least about 8%, at least about 9%, at least about 10%, at least about 15%, or at least about 20% compared to disease screening and / or diagnostic assays that do not utilize enrichment of cell-free microbial nucleic acid molecules and / or depletion of human nucleic acid molecules described elsewhere herein.

[0082] In some cases, the disclosure provided herein describes methods and / or systems configured to determine, identify, classify, and / or generate one or more microbial nucleic acid molecule features of one or more enriched microbial nucleic acid molecules from a biological sample of a subject, which can distinguish, classify, screen, and / or diagnose the health status of one or more subjects and / or one or more groups of subjects. In some cases, the biological sample can include a liquid biological sample, a tissue biological sample, or a combination thereof. In some cases, one or more nucleic acid molecule features of one or more subjects can be used to train a prediction model described elsewhere herein. In some cases, one or more nucleic acid molecule features can be derived from, obtained from, received from, and / or determined from one or more enriched microbial nucleic acid molecules of one or more biological samples of a subject and / or a plurality of subjects. In some examples, the microbial nucleic acid molecule can include one or more nucleic acid molecules from bacteria, fungi, viruses, or any combination thereof. In some cases, the health status of one or more subjects described elsewhere herein can include a cancerous health status, a non-cancerous disease health status, a healthy health status (i.e., when the subject does not have cancer or a non-cancerous disease). In some examples, the cancerous health status can include an individual having cancer. In some cases, the cancer can include cancer of the lung, breast, ovary, gastrointestinal tract, head and neck, liver, pancreas, prostate, skin, or any combination thereof. In some cases, lung cancer can include non-small cell lung cancer. In some cases, the cancerous health status can include a diagnosis of the stage of cancer (e.g., stage I, stage II, stage III, etc.). In some cases, the health status can include the spatial location (i.e., anatomical location) of cancer and / or disease within a subject or a plurality of subjects. In some cases, the health status can include the tissue and / or organ of origin of cancer. In some cases, the non-cancerous disease health status can include a lung disease. In some examples, the lung disease can include carcinoid, hamartoma, granuloma, interstitial fibrosis, emphysema, bronchitis, chronic obstructive pulmonary disease, pneumonia, sarcoidosis, or any combination thereof. In some cases, the liquid biological sample can include a liquid biopsy.In some cases, a liquid biopsy may include plasma, serum, whole blood, urine, cerebrospinal fluid, saliva, sweat, tears, exhaled condensate, or any combination thereof. In some cases, a tissue biopsy sample may include a tissue biopsy from one or more regions, organs, and / or anatomical locations of a subject (e.g., lung, skin, liver, pancreas, brain, etc.).

[0083] Method In some cases, the present disclosure provides a method 300 for enriching microbial nucleic acid molecules, as shown in FIG. 1. In some cases, the method comprises receiving, providing, and / or obtaining a biological sample (308, 306, 304, 312) containing one or more nucleic acid molecules (e.g., microbial 312 and / or human nucleic acid molecules) 302, depleting a biological sample of a protein (e.g., a histone protein) 308 coupled to one or more human nucleic acid molecules 304 by removing an affinity-based probe 314 coupled to a protein-human DNA complex (316), thereby enriching microbial nucleic acid molecules 312 of the biological sample 318. In some cases, one or more endonucleases 310 may be used to cleave or separate one or more segments of human nucleic acid molecules (e.g., DNA) bound to the protein(s). In some examples, the protein may include a transcription factor 306 that may include human DNA coupled to the transcription factor 306. In some cases, the biological sample may include a liquid biopsy sample, a tissue biopsy sample, or any combination thereof. In some cases, the liquid biopsy sample may include whole blood 320, plasma 322, serum, urine, cerebrospinal fluid, saliva, sweat, tears, exhaled condensate, or any combination, dilution, or processed fraction thereof. In some cases, plasma 322 may be obtained, isolated, and / or separated from whole blood 320 by centrifugation. In some cases, the microbial nucleic acid molecules 312 may include nucleic acid molecules derived from bacteria, viruses, fungi, or any combination thereof. In some examples, the microbial nucleic acid molecules may include cell-free microbial nucleic acid molecules (e.g., cell-free microbial DNA and / or RNA).

[0084] In some cases, the affinity-based probe can include an antibody, which can include a binding motif configured to couple to one or more regions and / or surfaces of a protein-human nucleic acid molecule (e.g., DNA) complex. In some cases, the binding motif of the antibody can couple to an epitope that includes human DNA and one or more histone proteins. In some cases, the antibody can include an anti-nucleosome antibody. In some examples, the antibody can bind to a solid support. The solid support can include magnetic beads, agarose beads, non-magnetic latex, functionalized sepharose, pH-sensitive polymers, or any combination thereof.

[0085] In some embodiments, as shown in FIG. 2A, provided herein is a method of depleting a biological sample of human nucleic acid molecules bound to nucleosomes using an affinity-based probe coupled to the surface of a solid support described elsewhere herein and coupled to the solid support 100. In some examples, the method can include providing, obtaining, and / or collecting a biological sample (e.g., a liquid biological sample), wherein the biological sample includes one or more microorganisms and mammalian nucleic acid molecules (step 101), exposing the biological sample to a solid support, wherein the surface of the solid support includes one or more nucleosome-specific affinity-based probes (step 102), and separating the solid support from the biological sample to remove bound nucleosomes coupled to a plurality of nucleosome-specific affinity-based probes, thereby enriching one or more microbial nucleic acid molecules of the biological sample (step 103). In some cases, the biological sample can include whole blood. In some examples, the solid support can be exposed to the plasma of a whole blood sample.

[0086] In some embodiments, as shown in FIG. 2B, provided herein is a method for depleting a biological sample of human nucleic acid molecules bound to nucleosomes using affinity-based probes configured to couple to a solid support 108. In some cases, the method comprises providing, obtaining, and / or collecting a biological sample comprising one or more microbial and mammalian nucleic acid molecules, wherein the mammalian nucleic acid molecules are bound to nucleosome(s), step 101 of providing, obtaining, and / or collecting; exposing the biological sample to an epitope and / or affinity tag-labeled nucleosome-specific affinity-based probe as described elsewhere herein, step 104; exposing the biological sample to one or more solid supports, wherein the surface of the one or more solid supports comprises a capture molecule configured to couple to one or more affinity-based probes, step 105; removing the solid support from the biological sample, thereby depleting the sample of one or more mammalian nucleic acid molecules bound to nucleosome(s), thereby enriching for one or more microbial nucleic acid molecules, step 106. In some cases, the capture molecule can comprise an anti-species antibody configured to couple and / or bind to a human and / or non-human mammalian (e.g., rabbit, mouse) antibody.

[0087] In some cases, as shown in FIG. 3A, the nucleic acid molecule library of enriched microbial nucleic acid molecules of the nucleosome-depleted biological sample 201 described elsewhere herein can be sequenced and / or analyzed to determine one or more microbial taxonomic features (e.g., microbial abundance and / or distribution of microbial classifications) using the microbial taxonomy workflow and / or method 114. In some cases, the microbial taxonomy method 114 includes sequencing the nucleic acid molecule library described elsewhere herein to generate a set of one or more nucleic acid molecule sequencing reads 108, filtering the set of one or more nucleic acid molecule sequencing reads 109 to remove one or more mammalian and / or human nucleic acid molecule sequencing reads, thereby generating a set of mammalian-depleted microbial sequencing reads 110, determining a microbial taxonomic assignment from the mammalian-depleted microbial sequencing reads 111, and / or purifying the microbial taxonomic assignment 112 to generate one or more purified microbial taxonomic feature sets 113. In some cases, filtering can include computationally mapping one or more nucleic acid sequencing reads to human reference genome data to identify and remove one or more mammalian sequencing reads of the one or more nucleic acid molecule sequencing reads. In some cases, purification can include in silico purification, experimental control purification, or a combination thereof. In some examples, one or more microbial taxonomic feature sets labeled with the health status of the subject from which the biological sample was received, obtained, and / or provided can be used to train the prediction models described elsewhere herein.

[0088] In some cases, experimental control purification may involve removing and / or subtracting microbial sequencing reads obtained from and / or received from nucleic acid molecule sequencing reads of one or more microbial nucleic acids collected and / or obtained from a subject biological sample from microbial sequencing reads obtained from and / or received from a control or blank nucleic acid molecule extraction kit (e.g., a sample collection container used to collect and / or provide one or more nucleic acid molecules of a biological sample). In some cases, the control and / or blank nucleic acid molecule extraction kit may not have a biological sample introduced into the kit. In some cases, a sample of the contents of the control and / or blank nucleic acid molecule extraction kit may be obtained and / or collected by swabbing and / or washing the control and / or blank nucleic acid molecule extraction kit and / or container with an eluate or buffer. In some cases, the nucleic acid molecule extraction kit may include a kit for extracting one or more microbial nucleic acid molecules of a biological sample described elsewhere herein. In some cases, removing and / or subtracting background or noise contaminating nucleic acid molecule sequencing reads (plural) (e.g., human and / or microbial nucleic acid molecules) present in a sample collection kit improves the classification, characterization, and / or diagnostic accuracy, sensitivity, and / or specificity of a model trained with one or more microbial features determined from purified microbial sequencing reads. In some cases, the improvement may include at least about 1%, at least about 2%, at least about 3%, at least about 4%, at least about 5%, at least about 6%, at least about 7%, at least about 8%, at least about 9%, at least about 10%, at least about 11%, at least about 12%, at least about 13%, at least about 14%, at least about 15%, or at least about 20% improvement in the performance characteristics of the accuracy, sensitivity, specificity, or any combination thereof of a prediction model described elsewhere herein.

[0089] In some cases, purification may include removing microbial contaminants (e.g., in silico purification) from identified microbial signatures (i.e., derived from one or more microbial nucleic acid molecules described elsewhere herein) prior to training the prediction model. In some cases, the microbes and their corresponding microbial nucleic acids can be removed based on statistical tests such as Fisher's exact test that account for differences in the proportionality of the presence of microbial nucleic acids between negative controls and biological samples. In some cases, the experimental control purification method can include: (i) obtaining one or more negative control containers (e.g., those of a nucleic acid molecule extraction kit) or chambers or reagents used to transport and / or store and / or process one or more biological samples; (ii) sequencing the nucleic acid molecules of the one or more negative control containers, thereby generating a plurality of negative control sequencing reads; (iii) mapping the plurality of negative control sequencing reads to a microbial genomic database, thereby generating a plurality of microbial nucleic acid molecule reads; and (iv) removing a plurality of negative control microbial nucleic acid molecule reads from the microbial nucleic acid molecule reads of the one or more biological samples prior to training the prediction model using one or more microbial signatures of the microbial nucleic acid molecule reads.

[0090] In some cases, the nucleic acid molecule library 107 can be generated and / or prepared from one or more enriched and / or amplified (described elsewhere herein) microbial nucleic acid molecules 104, and the nucleic acid molecule library 107 can be sequenced 108 (e.g., shotgun sequencing, next-generation sequencing, and / or synthetic sequencing) as described elsewhere herein. In some cases, the nucleic acid molecule library can include a single-stranded DNA nucleic acid molecule library.

[0091] In some examples, as shown in FIG. 3B, a nucleic acid molecule library of enriched microbial nucleic acid molecules of the nucleosome-depleted biological sample 201 described elsewhere herein can be sequenced and / or analyzed to determine one or more microbial functional features using the microbial function workflow and / or method 117. In some cases, the microbial function workflow and / or method 117 can include sequencing a nucleic acid molecule library to generate a set of one or more nucleic acid molecule sequencing reads 108, filtering a set of one or more nucleic acid molecule sequencing reads 109 to remove one or more mammalian and / or human nucleic acid molecule sequencing reads, thereby generating a set of mammalian-depleted microbial sequencing reads 110, determining a microbial taxonomic assignment from the mammalian-depleted microbial sequencing reads 111, purifying the microbial taxonomic assignment 112, and / or determining one or more microbial functional annotations 115 from those of the microbial taxonomic assignment to generate a set of one or more microbial functional features. In some cases, filtering can include computationally mapping one or more nucleic acid sequencing reads to human reference genome data to identify and remove one or more mammalian sequencing reads of the one or more nucleic acid molecule sequencing reads. In some cases, purification can include in silico purification. In some examples, one or more sets of microbial functional features labeled with the health state of a subject from whom a biological sample has been received, obtained, and / or provided can be used to train a prediction model described elsewhere herein.

[0092] In some embodiments, a biological sample 201 depleted of human nucleic acid molecules (e.g., human nucleic acid molecules coupled to nucleosomes) and / or enriched with microbial nucleic acid molecules can be enriched and amplified in one or more microbial nucleic acid amplification workflows 202, as shown in FIG. 5. In some cases, the microbial nucleic acid amplification workflow 202 can include hybridization probe enrichment 203, protein-microbial DNA enrichment 204, or a combination thereof. In some cases, the enriched microbial nucleic acid molecules can then be amplified by one or more marker gene amplification methods 205. In some cases, the biological sample amplified and / or further enriched after enrichment and / or amplification can be sequenced by targeted microbial amplicon sequencing (206), which can include shotgun sequencing, next-generation sequencing, sequencing by synthesis, or a combination thereof. In some examples, the marker gene amplification 205 can include forward and / or reverse primer polymerase chain reaction (PCR), inverse PCR, anchor PCR, primer-directed rolling circle amplification, or any combination thereof.

[0093] In some cases, hybridization probe enrichment may involve exposing, providing, and / or incubating one or more hybridization probes with a biological sample, and the one or more hybridization probes are configured to bind to non-mammalian nucleic acid molecules, such as microbial DNA, RNA, cell-free DNA, cell-free RNA, or any combination thereof. In some cases, the one or more hybridization probes may comprise nucleic acid molecules. In some cases, the one or more nucleic acid molecule hybridization probes may comprise sequences configured to hybridize to microbial nucleic acid molecule genomic features. In some cases, the microbial nucleic acid molecule genomic features may comprise one or more microbial genes or portions thereof. In some cases, the microbial nucleic acid molecule genomic features may comprise ribosomal RNA gene 5S, ribosomal RNA gene 16S, ribosomal RNA gene 23S, bacterial housekeeping gene dnaG, frr, infC, nusA, pgk, pyrG, rplA, rplB, rplC, rplD, rplE, rplF, rplK, rplL, rplM, rplN, rplP, rplS, rplT, rpmA, rpoB, rpsB, rpsC, rpsE, rpsI, rpsJ, rpsK, rpsM, rpsS, smpB, tsf, or any combination thereof. The microbial marker gene may comprise one or more fungal genes: ribosomal RNA gene 18S, ribosomal RNA gene 5.8S, ribosomal RNA gene 28S, and internal transcribed spacer regions 1 and 2.

[0094] In some examples, a method for enriching microbial nucleic acids with hybridization probes includes exposing, providing, and / or combining a nucleosome-depleted biological sample with one or more hybridization probes, incubating the hybridization probes and the nucleosome-depleted biological sample under conditions that promote nucleic acid base pairing (i.e., nucleic acid base hybridization) between the hybridization probes and one or more microbial nucleic acid molecules of the biological sample, and separating and / or removing unbound hybridization probes and hybridization probes bound to one or more microbial nucleic acid molecules of the biological sample, thereby enriching one or more microbial nucleic acid molecules of the biological sample. In some examples, the method may further include washing hybridization probes bound to one or more microbial nucleic acid molecules. In some examples, the washing may be configured to remove non-specifically associated nucleic acid molecules and other reaction components that may couple, hybridize, and / or bind to the hybridization probes.

[0095] In some cases, the nucleosome-depleted biological sample described elsewhere herein can be enriched by protein-based enrichment 204 configured to enrich one or more microbial nucleic acid molecules of the biological sample. In some cases, the method of protein-based enrichment comprises exposing, providing, and / or combining a nucleosome-depleted biological sample with one or more recombinant CXXC domain proteins to form a protein binding reaction, incubating the protein-DNA binding reaction under conditions that promote the interaction of the recombinant CXXC domain protein with the unmethylated CpG motifs of one or more microbial nucleic acid molecules of the nucleosome-depleted biological sample, and separating the unbound recombinant CXXC domain protein and the recombinant CXXC domain protein bound to the unmethylated CpG nucleic acid fragments from the remainder of the protein-DNA binding reaction, thereby enriching one or more microbial nucleic acid molecules. In some cases, the method of protein-based enrichment can further comprise washing the recombinant CXXC domain protein bound to the unmethylated CpG nucleic acid fragment to remove non-specifically associated nucleic acid molecules and the remainder of the protein-DNA binding reaction components.

[0096] In some examples, amplification of a marker gene can generate one or more microbial nucleic acid molecule amplicons. In some cases, one or more microbial nucleic acid molecule amplicons can include one or more genomic features. One or more genomic features can include a microbial phylogenetic marker gene or a marker gene fragment thereof. In some cases, the microbial phylogenetic marker gene can include marker genes for bacteria, fungi, or any combination thereof. The microbial phylogenetic marker gene can include a bacterial marker gene or a marker gene fragment thereof. The microbial phylogenetic marker gene can include a fungal marker gene or a marker gene fragment thereof. In some cases, the bacterial marker gene can include the ribosomal RNA gene 5S, ribosomal RNA gene 16S, ribosomal RNA gene 23S, the bacterial housekeeping gene dnaG, frr, infC, nusA, pgk, pyrG, rplA, rplB, rplC, rplD, rplE, rplF, rplK, rplL, rplM, rplN, rplP, rplS, rplT, rpmA, rpoB, rpsB, rpsC, rpsE, rpsI, rpsJ, rpsK, rpsM, rpsS, smpB, tsf, or any combination thereof. In some cases, the fungal marker gene can include the ribosomal RNA gene 18S, ribosomal RNA gene 5.8S, ribosomal RNA gene 28S, and one or more of the internal transcribed spacer regions 1 and 2.

[0097] In some cases, the enriched microbial nucleic acid molecules of the nucleosome-depleted biological sample 201 can be amplified by one or more microbial nucleic acid amplification workflows 202, as seen in FIG. 6A, and analyzed and / or processed by a workflow and / or method 213 to generate one or more microbial taxonomic feature sets 212. In some cases, the workflow and / or method sequences one or more microbial amplicons generated by the amplification described elsewhere herein (206), thereby generating one or more sets of microbial sequencing reads (207), filters one or more sets of microbial sequencing reads to remove one or more microbial mitochondrial nucleic acid molecules (e.g., microbial mitochondrial DNA) 208, generating one or more sets of microbial mitochondrial-depleted microbial nucleic acid sequencing reads 209, assigns and / or determines the microbial taxonomy of one or more sets of microbial mitochondrial-depleted microbial nucleic acid sequencing reads 210, purifies the assigned and / or determined microbial taxonomy 211, determines the microbial functional annotation 214 of the purified microbial taxonomy, and / or generates one or more microbial functional feature sets 215. In some cases, filtering one or more sets of microbial sequencing reads to remove one or more microbial mitochondrial nucleic acid molecules can include computationally mapping one or more microbial nucleic acid sequencing reads to mitochondrial reference genomic data to identify and remove one or more mitochondrial DNA sequencing reads. In some cases, purification can include in silico purification. In some examples, one or more microbial taxonomic feature sets labeled with the health status of the subject from which the biological sample was received, obtained, and / or provided can be used to train the prediction models described elsewhere herein.

[0098] In some cases, the enriched microbial nucleic acid molecules of the nucleosome-depleted biological sample 201 can be amplified by one or more microbial nucleic acid amplification workflows 202, as seen in FIG. 6B, and analyzed and / or processed by a workflow and / or method 216 to generate one or more microbial functional feature sets 215. In some cases, the workflow and / or method sequences one or more microbial amplicons generated by the amplification described elsewhere herein (206), thereby generating one or more sets of microbial sequencing reads (207), filters one or more sets of microbial sequencing reads to remove one or more microbial mitochondrial nucleic acid molecules (e.g., microbial mitochondrial DNA) 208, generating one or more sets of microbial mitochondrial-depleted microbial nucleic acid sequencing reads 209, assigns and / or determines the microbial classification of one or more sets of microbial mitochondrial-depleted microbial nucleic acid sequencing reads 210, purifies the assigned and / or determined microbial classification 211, and may include generating one or more microbial functional feature sets 212. In some cases, filtering one or more sets of microbial sequencing reads to remove one or more microbial mitochondrial nucleic acid molecules may include computationally mapping one or more microbial nucleic acid sequencing reads to mitochondrial reference genomic data to identify and remove one or more mitochondrial DNA sequencing reads. In some cases, purification may include in silico purification. In some examples, one or more microbial taxonomic feature sets labeled with the health status of the subject from which the biological sample was received, obtained, and / or provided can be used to train the prediction models described elsewhere herein.

[0099] In some cases, the enriched microbial nucleic acid molecules of the nucleosome-depleted biological sample 201 can be amplified by one or more microbial nucleic acid amplification workflows 202, as seen in FIG. 6C, and analyzed and / or processed by workflow and / or method 221 to generate one or more microbial amplicon sequence variant (ASV) feature sets 220. In some cases, the workflow and / or method sequences one or more microbial amplicons generated by amplification as described elsewhere herein (206), thereby generating one or more sets of microbial sequencing reads (207), filters one or more sets of microbial sequencing reads to remove one or more microbial mitochondrial nucleic acid molecules (e.g., microbial mitochondrial DNA) 208, generating one or more sets of microbial mitochondrial-depleted microbial nucleic acid sequencing reads 209, identifying, assigning, and / or determining the ASV features of one or more sets of microbial mitochondrial-depleted microbial nucleic acid sequencing reads 217, enumerating the ASV features 218, purifying the enumerated ASV features (e.g., in silico purification as described elsewhere herein) 219, and producing one or more purified microbial ASV feature sets 220. In some cases, identifying the ASV features of one or more sets of microbial mitochondrial-depleted nucleic acid sequencing reads may include identifying mutations and / or single nucleotide polymorphisms of one or more microbial genes. In some cases, sequence variations that result in mutations and / or single nucleotide polymorphisms of one or more microbial genes may provide a measure of the microbial diversity of the biological sample of interest. In some cases, microbial diversity can be utilized to determine one or more microbial features used in the training of predictive models as described elsewhere herein. In some examples, the enumeration of ASV features may include determining the count or frequency (e.g., histogram) of microbial gene variants, mutations, and / or single nucleotide polymorphisms (i.e., ASV features) of one or more microbial genes.In some cases, filtering one or more sets of microbial sequencing reads to remove one or more microbial mitochondrial nucleic acid molecules may involve computationally mapping one or more microbial nucleic acid sequencing reads to mitochondrial reference genomic data to identify and remove one or more mitochondrial DNA sequencing reads. In some cases, the purification may include in silico purification. In some examples, one or more sets of microbial ASV features labeled with the health status of a subject from which a biological sample has been received, obtained, and / or provided can be used to train a predictive model described elsewhere herein.

[0100] Predictive model The methods and systems described herein utilize or access external capabilities of artificial intelligence, predictive models, and / or machine learning trained on one or more sets of microbial nucleic acid features, such as microbial taxonomic features, microbial functional features, microbial ASV features, or any combination thereof, that can classify, diagnose, and / or characterize the health status of a subject, multiple subjects, and / or one or more groups of subjects. In some cases, one or more microbial nucleic acid molecule features (e.g., microbial functional features, microbial taxonomic features, etc.) described elsewhere herein can predict, classify, and / or identify cancer and / or non-cancerous diseases in one or more subjects. In some cases, one or more microbial nucleic acid molecule features can be used to train one or more predictive models described elsewhere herein. These trained predictive models can be used to accurately predict, classify, and / or characterize the health status of a subject, multiple subjects, and / or one or more groups of subjects, such as cancer, non-cancerous diseases, disorders, or any combination thereof. Using such predictive capabilities, a healthcare provider (e.g., a physician) can make informed, accurate risk-based decisions, thereby improving the quality of care and monitoring provided to subjects having cancer, non-cancerous diseases, disorders, or any combination thereof.

[0101] The methods and systems of the present disclosure can analyze the presence and / or abundance of microorganisms (e.g., the abundance, classification, microbial functional pathways of microorganisms of a particular genus). Then, using the presence and / or abundance of the microorganisms, one or more nucleic acid molecule features, e.g., non-mammalian and / or microbial nucleic acid molecule features, that can predict one or more cancers and / or non-cancerous diseases of a subject can be determined. In some cases, the methods and systems described elsewhere herein can train a prediction model using one or more nucleic acid molecule features indicative of a subject's health status, e.g., cancer and / or non-cancerous diseases. In some cases, the trained prediction model can then be used to generate the likelihood (e.g., prediction) of cancer and / or non-cancerous diseases of one or more subjects different from the one or more subjects used to train the prediction model. The trained prediction model can include an artificial intelligence-based model, such as a machine learning-based classifier, configured to process one or more nucleic acid molecule features from one or more nucleic acid molecules and / or one or more enriched, filtered, and / or amplified nucleic acid molecules to generate the likelihood that a subject(s) has cancer, a non-cancerous disease, or a disorder. The model can be trained using the abundance of microbial taxonomic features or microbial functional pathways from one or more cohorts of subjects, e.g., cancer subjects, subjects with non-cancerous diseases, subjects without diseases and without cancer, cancer subjects receiving treatment for cancer, subjects receiving treatment for non-cancerous diseases, or any combination thereof. In some cases, the prediction model can be trained to provide a treatment prediction for treating the cancer of one or more subjects that are not part of the training dataset of the prediction model. Such a prediction model can output treatment recommendations for one or more subjects that are not part of the training dataset when an input of the presence and abundance of one or more microorganisms in a patient's hybridization-enriched biological sample is provided.

[0102] In some embodiments, the present disclosure provides a method and / or system for generating one or more classifiers from one or more microbial nucleic acid features of one or more subjects identified and / or determined from nucleosome-depleted biological samples of the one or more subjects, as described elsewhere herein.

[0103] In some cases, the method and / or system may include training a prediction model with one or more microbiological taxonomic features identified and / or determined from one or more nucleosome-depleted biological samples of one or more subjects, as shown in FIG. 4A. In some examples, the method provides, receives, and / or obtains nucleosome-depleted biological samples described elsewhere herein from one or more subjects classified and / or characterized as (e.g., by a gold standard diagnostic and / or classification method) healthy, having cancer, and / or having a non-cancerous disease, generating a nucleic acid molecule (e.g., DNA) sequencing library from one or more microbial nucleic acids of the depleted biological sample, sequencing and / or analyzing one or more microbial nucleic acid molecules with the microbial taxonomic methods and / or workflows described elsewhere herein, training a prediction model (e.g., a machine learning classifier) with one or more microbiological taxonomic features determined, identified, and / or analyzed from the microbial nucleic acid molecule sequencing reads, thereby generating a trained prediction model (e.g., a diagnostic model) including a classifier for a healthy-versus-cancer classifier, a cancer-versus-non-cancerous disease classifier, a non-cancerous disease-versus-healthy classifier, or any combination thereof. In some cases, the trained prediction model and / or one or more classifiers (123, 124, 125) may be used to screen, diagnose, and determine the health status of one or more subjects not included in the training of the prediction model by providing one or more microbiological taxonomic features of the nucleosome-depleted biological samples of the one or more subjects. In some cases, the one or more microbiological taxonomic features of the nucleosome-depleted biological samples of the one or more subjects may be determined by the methods and systems described elsewhere herein.

[0104] In some cases, the method and / or system may include training a prediction model with one or more microbial functional features identified and / or determined from one or more nucleosome-depleted biological samples of one or more subjects, as shown in Figure 4B. In some examples, the method provides, receives, and / or obtains a nucleosome-depleted biological sample as described elsewhere herein from one or more subjects classified as and / or characterized as healthy, having cancer, and / or having a non-cancerous disease (e.g., by a gold standard diagnostic and / or classification method), generating a nucleic acid molecule (e.g., DNA) sequencing library from one or more microbial nucleic acids of the depleted biological sample, sequencing and / or analyzing one or more microbial nucleic acid molecules with the microbial functional methods and / or workflows described elsewhere herein, training a prediction model (e.g., a machine learning classifier) with one or more microbial functional features determined, identified, and / or analyzed from the microbial nucleic acid molecule sequencing reads, thereby generating a trained prediction model (e.g., a diagnostic model) including a classifier for a healthy vs. cancer classifier, a cancer vs. non-cancerous disease classifier, a non-cancerous disease vs. healthy classifier, or any combination thereof. In some cases, the trained prediction model and / or one or more classifiers (123, 124, 125) may be used to screen, diagnose, and determine the health status of one or more subjects not included in the training of the prediction model by providing one or more microbial functional features of the nucleosome-depleted biological samples of the one or more subjects. In some cases, the one or more microbial functional features of the nucleosome-depleted biological samples of the one or more subjects may be determined by the methods and systems described elsewhere herein.

[0105] The prediction model and / or the trained prediction model may include one or more prediction models. The model may include one or more machine learning algorithms. Examples of machine learning algorithms may include support vector machine (SVM), naive Bayes classification, random forest, neural network (such as deep neural network (DNN)), regression neural network (RNN), deep RNN, long short-term memory (LSTM) regression neural network (RNN), gated recurrent unit (GRU), gradient boosting machine, random forest, or other supervised learning algorithms or unsupervised machine learning, statistics, linear regression, k-nearest neighbor, k-means, decision tree, logistic regression, or any combination thereof. The model can be used for classification or regression. The model may similarly involve the estimation of an ensemble model consisting of multiple prediction models. For example, in the construction of gradient boosting decision trees, techniques such as gradient boosting may be utilized. The model can be trained using one or more nucleic acid molecule features, target data, such as the medical history of the subject, the medical history of the subject's family, the vital signs of the subject (e.g., blood pressure, pulse, body temperature, oxygen saturation), or any combination thereof in one or more training data sets.

[0106] The prediction model may include any number of machine learning algorithms. In some embodiments, the random forest machine learning algorithm may be an ensemble of bagged decision trees. The ensemble can be at least about 1, 2, 3, 4, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 120, 140, 160, 180, 200, 250, 500, 1000 or more bagged decision trees. The ensemble can be at most about 1000, 500, 250, 200, 180, 160, 140, 120, 100, 90, 80, 70, 60, 50, 40, 30, 20, 10, 5, 4, 3, 2 or fewer bagged decision trees. The ensemble can be about 1 - 1000, 1 - 500, 1 - 200, 1 - 100, or 1 - 10 bagged decision trees.

[0107] In some embodiments, the machine learning algorithm may have various parameters. The various parameters can be, for example, a learning rate, a mini-batch size, the number of epochs to train, momentum, learning weight decay, or neural network layers, etc.

[0108] In some embodiments, the learning rate can be from about 0.00001 to 0.1.

[0109] In some embodiments, the mini-batch size can be from about 16 to 128.

[0110] In some embodiments, the neural network may include neural network layers. The neural network can have at least about 2 to 1000 or more neural network layers.

[0111] In some embodiments, the number of epochs to train can be at least about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100, 150, 200, 250, 500, 1000, 10000, or more.

[0112] In some embodiments, the momentum can be at least about 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9 or more. In some embodiments, the momentum can be at most about 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, or less.

[0113] In some embodiments, the learning weight decay can be at least about 0.00001, 0.0001, 0.001, 0.002, 0.003, 0.004, 0.005, 0.006, 0.007, 0.008, 0.009, 0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, 0.1, or more. In some embodiments, the learning weight decay can be at most about 0.1, 0.09, 0.08, 0.07, 0.06, 0.05, 0.04, 0.03, 0.02, 0.01, 0.009, 0.008, 0.007, 0.006, 0.005, 0.004, 0.003, 0.002, 0.001, 0.0001, 0.00001, or less.

[0114] In some embodiments, the machine learning algorithm may use a loss function. The loss function can be, for example, regression loss, mean absolute error, mean bias error, hinge loss, Adam optimizer, and / or cross entropy.

[0115] In some embodiments, the parameters of the machine learning algorithm can be adjusted with the help of a human and / or a computer system.

[0116] In some embodiments, the machine learning algorithm may prioritize certain features. The machine learning algorithm may prioritize features that may be associated by detecting cancer, non-cancerous diseases, disorders, or any combination thereof. If a feature is classified more frequently than another feature in determining cancer, non-cancerous diseases, and / or disorders, that feature may be associated by detecting cancer, non-cancerous diseases, and / or disorders. In some cases, features can be prioritized using a weighting system. In some cases, features can be statistically prioritized based on the frequency and / or amount of occurrence of that feature. The machine learning algorithm may prioritize features with the help of a human and / or a computer system.

[0117] In some cases, the machine learning algorithm may prioritize certain features for reasons such as reducing computational cost, saving processing power, saving processing time, improving reliability, or reducing random access memory usage.

[0118] The training dataset can be generated, for example, from one or more cohorts of subjects having a diagnosis of a common cancer, non-cancerous disease, or disorder. The training dataset can include one or more nucleic acid molecule features in the form of taxonomic assignment features of the abundance of microorganisms present in a biological sample of one or more subjects and / or microbial functional pathway features of the microorganisms present in the biological sample. The features can include a cancer diagnosis of one or more subjects corresponding to the microbial features. In some cases, the features can include patient information such as the patient's age, the patient's medical history, other medical conditions, current or past medication treatments, clinical risk scores, and time since last observation. For example, a set of features collected from a given patient at a given time point can function collectively as a signature that can indicate the health state or status of the patient at that given time point.

[0119] The label can include, for example, clinical outcomes such as the presence, absence, diagnosis, and / or prognosis of cancer, non-cancerous disease, disorder, or combinations thereof in a subject (e.g., a patient). The clinical outcome can include treatment efficacy (e.g., whether the subject is a positive responder or a negative responder to cancer and / or disease-based treatment).

[0120] The input features can be structured by aggregating the data into bins or alternatively by using one-hot encoding. The input can also include feature values or vectors derived from the aforementioned inputs such as cross-correlation.

[0121] The training dataset can be constructed from the presence and / or abundance of one or more nucleic acid mole features of, for example, one or more microbial taxonomic features, one or more microbial functional pathways, or combinations thereof, identified and / or classified from enriched and / or amplified nucleic acid molecules of biological samples indicative of cancer, non-cancerous disease, disorder, or any combination thereof.

[0122] The model can process input features to generate output values including one or more classifications, one or more predictions, or combinations thereof. For example, such classifications or predictions may include a binary classification of the presence or absence of cancer, the presence of a non-cancerous disease, the presence of a disorder, or any combination of those classifications of a subject. In some cases, one or more prediction models and / or machine learning algorithms can classify a subject between a group of categorical labels (e.g., "no cancer, non-cancerous disease and / or disorder", "apparent cancer, non-cancerous disease and / or disorder", and "possible cancer, non-cancerous disease and / or disorder"), the likelihood (e.g., relative likelihood or probability) of developing a particular cancer, non-cancerous disease, and / or disorder, "risk factors" regarding the presence of cancer, non-cancerous disease and / or disorder, the likelihood of patient death, and a score indicating a confidence interval for any numerical prediction. Various machine learning techniques may be cascaded such that the output of a machine learning technique can also be used as input features to subsequent layers or subsections of the model.

[0123] To train a model for generating real-time classifications or predictions (e.g., by determining the weights and correlations of the model), the training datasets and / or one or more training features described elsewhere in this specification can be used to train the model. Such datasets and / or features can be of a size sufficient to generate statistically significant classifications or predictions. For example, a dataset can include one or more nucleic acid molecule features derived from sequencing data of the presence and / or abundance of microorganisms, such as fungi, viruses, archaea, bacteria, or any combination thereof, in a biological sample of one or more subjects.

[0124] The dataset can be split into subsets (e.g., individual or overlapping), such as a training dataset, a development dataset, and a test dataset. For example, the dataset can be split into a training dataset that includes 80% of the dataset, a development dataset that includes 10% of the dataset, and a test dataset that includes 10% of the dataset. The training dataset can include about 10%, about 20%, about 30%, about 40%, about 50%, about 60%, about 70%, about 80%, or about 90% of the dataset. The development dataset can include about 10%, about 20%, about 30%, about 40%, about 50%, about 60%, about 70%, about 80%, or about 90% of the dataset. The test dataset can include about 10%, about 20%, about 30%, about 40%, about 50%, about 60%, about 70%, about 80%, or about 90% of the dataset. In some embodiments, leave-one-out cross-validation can be used. The training set (e.g., the training dataset) can be selected by random sampling of a set of data corresponding to one or more patient cohorts to ensure the independence of sampling. Alternatively, the training set (e.g., the training dataset) can be selected by proportional sampling of a set of data corresponding to one or more patient cohorts to ensure the independence of sampling.

[0125] To improve the accuracy of model prediction and reduce overfitting of the model, the dataset can be augmented to increase the number of samples in the training set. For example, data augmentation can include rearranging the order of observations in the training records. To handle datasets with missing observations, methods for filling in missing data, such as forward filling, backward filling, linear interpolation, and multi-task Gaussian processes, can be used. The dataset can be filtered or batch-corrected to remove or mitigate confounding factors. For example, within a database, a subset of subjects can be excluded.

[0126] The model may include one or more neural networks such as a neural network, a convolutional neural network (CNN), a deep neural network (DNN), a recurrent neural network (RNN), or a deep RNN. The recurrent neural network may include units that can be long short-term memory (LSTM) units or gated recurrent units (GRUs). For example, the model may include an algorithm architecture that includes a neural network having an input feature set, such as one or more nucleic acid molecule features, vital measurements, the medical history of the subject, the demographics of the subject, or any combination thereof, as described elsewhere herein. Neural network techniques such as dropout or regularization may be used during training of the model to prevent overfitting. The neural network may include a plurality of sub-networks, each of which is configured to generate a classification or prediction of a different type of output information, which may be combined to form the overall output of the neural network. The machine learning model may alternatively utilize statistical or related algorithms including random forest, classification and regression trees, support vector machines, discriminant analysis, regression techniques, and their ensembles and gradient-boosted variations.

[0127] When the model generates a classification or prediction of cancer, non-cancerous disease, disorder, or a combination thereof, a notification (e.g., an alert or alarm) may be generated and transmitted to a healthcare provider such as a physician, nurse, or other member of the treatment team of the subject within the hospital. The notification may be transmitted via an automated phone call, a short message service (SMS), a multimedia message service (MMS) message, an email, and / or an alert within a dashboard. The notification may include output information such as a prediction of cancer, non-cancerous disease, and / or disorder; the likelihood of the predicted cancer, non-cancerous disease, and / or disorder; the time until the expected onset of the cancer, non-cancerous disease, and / or disorder; the confidence interval of the likelihood or time, the recommended treatment course for the cancer, non-cancerous disease, and / or disorder, or any combination of such information.

[0128] To verify the performance of a model, different performance metrics may be generated. For example, the area under the receiver operating characteristic curve (AUROC) can be used to determine the diagnostic, prognostic, screening, or any combination of those capabilities of the model. For example, the model may use an adjustable classification threshold such that the specificity and sensitivity are adjustable, and the receiver operating characteristic curve (ROC) can be used to identify different operating points corresponding to different values of specificity and sensitivity.

[0129] In some cases, such as when the dataset is not large enough, cross-validation can be performed to evaluate the robustness of the model across different training and test datasets.

[0130] The following definitions can be used to calculate performance metrics such as sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), area under the precision-recall curve (AUPR), AUROC, or the like. "False positive" can refer to an outcome where a positive outcome or result is generated incorrectly or prematurely (e.g., before the actual onset or without the onset of cancer, non-cancerous disease, and / or disorder). "True positive" can refer to an outcome where a positive outcome or result is correctly generated when the patient has cancer, non-cancerous disease, and / or disorder (e.g., the patient exhibits symptoms of cancer, non-cancerous disease, and / or disorder, or the patient's record indicates cancer, non-cancerous disease, and / or disorder). "False negative" can refer to an outcome where a negative outcome or result is generated, but the patient has cancer, non-cancerous disease, and / or disorder (e.g., the patient exhibits symptoms of cancer, non-cancerous disease, and / or disorder, or the patient's record indicates cancer, non-cancerous disease, and / or disorder). "True negative" can refer to an outcome where a negative outcome or result is generated (e.g., before the actual onset or without the onset of cancer, non-cancerous disease, and / or disorder).

[0131] The model may be trained until certain predetermined conditions for accuracy or performance are met, such as having a minimum desired value corresponding to a diagnostic accuracy metric. For example, the diagnostic accuracy metric may correspond to predicting the likelihood of the occurrence of cancer, non-cancerous diseases, and / or disorders in a subject. As another example, the diagnostic accuracy metric may correspond to predicting the likelihood of deterioration or recurrence of cancer, non-cancerous diseases, and / or disorders that the subject has been previously treated for. Examples of diagnostic accuracy metrics may include sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, AUPR, and AUROC corresponding to the diagnostic accuracy of detecting or predicting cancer, non-cancerous diseases, and / or disorders.

[0132] For example, such a predetermined condition may be that the sensitivity for predicting cancer, non-cancerous diseases, and / or disorders is, for example, at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0133] As another example, such a predetermined condition may be that the specificity for predicting cancer, non-cancerous diseases, and / or disorders is, for example, at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0134] As another example, such a predetermined condition may be that the positive predictive value (PPV) for predicting cancer, non-cancerous diseases, and / or disorders is, for example, at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0135] As another example, such predetermined conditions may be that the negative predictive value (NPV) for predicting cancer, non-cancerous diseases and / or disorders is, for example, at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0136] As another example, such predetermined conditions may be that the area under the curve (AUC) (AUROC) of the receiver operating characteristic curve (ROC) for predicting cancer, non-cancerous diseases and / or disorders is at least about 0.50, at least about 0.55, at least about 0.60, at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.85, at least about 0.90, at least about 0.95, at least about 0.96, at least about 0.97, at least about 0.98, or at least about 0.99.

[0137] As another example, such predetermined conditions may be that the area under the precision-recall curve (AUPR) for predicting cancer, non-cancerous diseases and / or disorders is at least about 0.10, at least about 0.15, at least about 0.20, at least about 0.25, at least about 0.30, at least about 0.35, at least about 0.40, at least about 0.45, at least about 0.50, at least about 0.55, at least about 0.60, at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.85, at least about 0.90, at least about 0.95, at least about 0.96, at least about 0.97, at least about 0.98, or at least about 0.99.

[0138] In some embodiments, the trained model can be trained or configured to predict cancer, non-cancerous diseases, and / or disorders with a sensitivity of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0139] In some embodiments, the trained model can be trained or configured to predict cancer, non-cancerous diseases, and / or disorders with a specificity of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0140] In some embodiments, the trained model can be trained or configured to predict cancer, non-cancerous diseases, and / or disorders with a positive predictive value (PPV) of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0141] In some embodiments, the trained model can be trained or configured to predict cancer, non-cancerous diseases, and / or disorders with a negative predictive value (NPV) of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%.

[0142] In some embodiments, the trained model can be trained or configured to predict cancer, non-cancerous diseases, and / or disorders with an area under the receiver operating characteristic curve (ROC) (AUROC) of at least about 0.50, at least about 0.55, at least about 0.60, at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.85, at least about 0.90, at least about 0.95, at least about 0.96, at least about 0.97, at least about 0.98, or at least about 0.99.

[0143] In some embodiments, the trained model can be trained or configured to predict cancer, non-cancerous diseases, and / or disorders with an area under the precision-recall curve (AUPR) of at least about 0.10, at least about 0.15, at least about 0.20, at least about 0.25, at least about 0.30, at least about 0.35, at least about 0.40, at least about 0.45, at least about 0.50, at least about 0.55, at least about 0.60, at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.85, at least about 0.90, at least about 0.95, at least about 0.96, at least about 0.97, at least about 0.98, or at least about 0.99.

[0144] The training dataset may be collected from a subject (e.g., a human). Each training has a diagnostic status indicating whether they have been diagnosed with a biological condition or have not been diagnosed with cancer, non-cancerous diseases, and / or disorders.

[0145] In some embodiments, the model is a neural network or a convolutional neural network. See Vincent et al., 2010, “Stacked denoising autoencoders: Learning useful representations in a deep network with a local denoising criterion,” J Mach Learn Res 11, pp. 3371-3408, Larochelle et al., 2009, “Exploring strategies for training deep neural networks,” J Mach Learn Res 10, pp. 1-40, and Hassoun, 1995, Fundamentals of Artificial Neural Networks, Massachusetts Institute of Technology. Each of these is incorporated herein by reference.

[0146] In some embodiments, independent component analysis (ICA) is used to de-dimensionalize the data, such as that described in Lee, T.-W. (1998): Independent component analysis: Theory and applications, Boston, Mass: Kluwer Academic Publishers, ISBN 0-7923-8261-7, and Hyvaerinen, A.; Karhunen, J.; Oja, E. (2001): Independent Component Analysis, New York: Wiley, ISBN 978-0-471-40540-5, which are hereby incorporated by reference in their entirety.

[0147] In some embodiments, principal component analysis (PCA) is used to reduce the dimensionality of the data, such as that described in Jolliffe, I.T. (2002). Principal Component Analysis. Springer Series in Statistics. New York: Springer-Verlag. doi: 10.1007 / b98835. ISBN 978-0-387-95442-4, which is hereby incorporated by reference in its entirety.

[0148] SVM is described in Cristianini and Shawe-Taylor, 2000, “An Introduction to Support Vector Machines,” Cambridge University Press, Cambridge; Boser et al., 1992, “A training algorithm for optimal margin classifiers,” in Proceedings of the 5th Annual ACM Workshop on Computational Learning Theory, ACM Press, Pittsburgh, Pa., pp. 142-152; Vapnik, 1998, Statistical Learning Theory, Wiley, New York; Mount, 2001, Bioinformatics: sequence and genome analysis, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.; Duda, Pattern Classification, Second Edition, 2001, John Wiley & Sons, Inc., pp. 259, 262-265; and Hastie, 2001, The Elements of Statistical Learning, Springer, New York; and Furey et al., 2000, Bioinformatics 16, 906-914, each of which is hereby incorporated by reference in its entirety. When used for classification, SVMs separate a given set of binary-labeled data from labeled data using a hyperplane that is maximally distant. If linear separation is not possible, SVMs can function in combination with a “kernel” technique that automatically performs a non-linear mapping into the feature space. The hyperplane found by the SVM in the feature space corresponds to a non-linear decision boundary in the input space.

[0149] Decision trees are outlined by Duda, 2001, Pattern Classification, John Wiley & Sons, Inc., New York, pp. 395-396, which is incorporated herein by reference. Decision tree-based methods partition the feature space into a set of rectangles and fit a model (such as a constant) to each. In some embodiments, the decision tree is a random forest regression. One particular algorithm that may be used is Classification and Regression Trees (CART). Other particular decision tree algorithms include, but are not limited to, ID3, C4.5, MART, and random forests. CART, ID3, and C4.5 are described in Duda, 2001, Pattern Classification, John Wiley & Sons, Inc., New York, pp. 396-408 and pp. 411-412, which is incorporated herein by reference. CART, MART, and C4.5 are described in Hastie et al., 2001, The Elements of Statistical Learning, Springer-Verlag, New York, Chapter 9, which is incorporated herein by reference in its entirety. Random forests are described in Breiman, 1999, “Random Forests - Random Features,” Technical Report 567, Statistics Department, U.C. Berkeley, September 1999, which is incorporated herein by reference in its entirety.

[0150] Clustering (e.g., unsupervised clustering model algorithms and supervised clustering model algorithms) is described on pages 211 - 256 of Duda and Hart, Pattern Classification and Scene Analysis, 1973, John Wiley & Sons, Inc., New York (hereinafter "Duda 1973"), which is hereby incorporated by reference in its entirety. As described in Section 6.7 of Duda 1973, the clustering problem is described as one of finding natural groupings in a dataset. To identify natural groupings, two problems are addressed. First, determine a method for measuring the similarity (or dissimilarity) between two samples. This metric (similarity measure) is used to ensure that samples within one cluster are more similar to each other than samples within other clusters. Second, determine a mechanism for dividing the data into clusters using the similarity measure. Similarity measures are discussed in Section 6.7 of Duda 1973, and it is described that one way to start a clustering investigation is to define a distance function and compute a matrix of distances between all pairs of samples in the training set. If distance is a good measure of similarity, the distances between reference entities within the same cluster will be significantly smaller than the distances between reference entities in different clusters. However, as described on page 215 of Duda 1973, clustering does not require the use of a distance metric. For example, a non - metric similarity function s(x,x’) can be used to compare two vectors x and x’. Conventionally, s(x,x’) is a symmetric function that has a large value when x and x’ are "similar" in some way. Examples of non - metric similarity functions s(x,x’) are shown on page 218 of Duda 1973. Once a method for measuring "similarity" or "dissimilarity" between points in a dataset is selected, clustering requires a criterion function to measure the clustering quality of any partition of the data. The data is clustered using a partition of the dataset that extremizes the criterion function. See page 217 of Duda 1973. Criterion functions are discussed in Section 6.8 of Duda 1973.More recently, Duda et al., Pattern Classification, 2nd edition, John Wiley & Sons, Inc. New York has been published. Pages 537 - 563 describe clustering in detail. Details of clustering techniques can be found in Kaufman and Rousseeuw, 1990, Finding Groups in Data: An Introduction to Cluster Analysis, Wiley, New York, N.Y., Everitt, 1993, Cluster analysis (3d ed.), Wiley, New York, N.Y., and Backer, 1995, Computer - Assisted Reasoning in Cluster Analysis, Prentice Hall, Upper Saddle River, New Jersey, each of which is incorporated herein by reference. Specific exemplary clustering techniques that may be used in the present disclosure include hierarchical clustering (agglomerative clustering using the nearest neighbor algorithm, farthest neighbor algorithm, average linkage algorithm, centroid algorithm, or sum - of - squares algorithm), k - means clustering, fuzzy k - means clustering, and Jarvis - Patrick clustering, but are not limited thereto. In some embodiments, clustering includes unsupervised clustering, where no preconception is imposed as to which clusters should be formed when the training set is being clustered.

[0151] Regression models, such as those of the multiple-category logit model, are described in Agresti, An Introduction to Categorical Data Analysis, 1996, John Wiley & Sons, Inc., New York, Chapter 8, which is hereby incorporated by reference in its entirety. In some embodiments, the model utilizes the regression models disclosed in Hastie et al., 2001, The Elements of Statistical Learning, Springer-Verlag, New York, which is hereby incorporated by reference in its entirety. In some embodiments, gradient boosting models are used, for example, for the classification algorithms described herein, and these gradient boosting models are described in Boehmke, Bradley; Greenwell, Brandon (2019). “Gradient Boosting”. Hands-On Machine Learning with R. Chapman & Hall. pp. 221-245. ISBN 978-1-138-49568-5., which is hereby incorporated by reference in its entirety. In some embodiments, ensemble modeling techniques are used, and these ensemble modeling techniques are described in the implementation of the classification model herein and in Zhou Zhihua (2012) Ensemble Methods: Foundations and Algorithms. Chapman and Hall / CRC. ISBN 978-1-439-83003-1 (which is hereby incorporated by reference in its entirety).

[0152] In some embodiments, the machine learning analysis is performed by a device that executes one or more programs (e.g., one or more programs stored in non-persistent memory or persistent memory) that include instructions for performing data analysis. In some embodiments, the data analysis is performed by a system that includes at least one processor (e.g., a processing core) and a memory (e.g., one or more programs stored in non-persistent memory or persistent memory) that includes instructions for performing data analysis.

[0153] Computer system The present disclosure provides a computer system programmed to implement the methods of the present disclosure. FIG. 7 shows a computer system 600 programmed or otherwise configured to predict the health state of one or more target cancers, non-cancerous diseases, or any combination thereof, train a prediction model described elsewhere herein, generate a recommended treatment, or perform any combination of the methods described elsewhere herein. The computer system 600 can be a user's electronic device or a computer system located remotely with respect to the electronic device. The electronic device can be a mobile electronic device.

[0154] The computer system 600 includes a central processing unit (CPU, also referred to herein as "processor" and "computer processor") 606, which can be a single-core or multi-core processor, or multiple processors for parallel processing. The computer system 600 also includes a memory or memory location 604 (e.g., random access memory, read-only memory, flash memory), an electronic storage unit 602 (e.g., hard disk), a communication interface 608 (e.g., network adapter) for communicating with one or more other systems, and peripheral devices 610 such as caches, other memories, data storage devices, and / or electronic display adapters. The memory 604, storage unit 602, interface 608, and peripheral devices 610 communicate with the CPU 606 via a communication bus (solid line), such as a motherboard. The storage unit 602 can be a data storage unit (or data storage location) for storing data. The computer system 600 can be operably coupled to a computer network ("network") 612 using the communication interface 608. The network 612 can be the Internet, the Internet and / or an extranet, or an intranet and / or an extranet communicating with the Internet. In some cases, the network 612 is a telecommunications network and / or a data network. The network 612 can include one or more computer servers that enable distributed computing such as cloud computing. The network 612 can implement a peer-to-peer network in some cases where the computer system 600 is used, which can enable devices coupled to the computer system 600 to operate as clients or servers.

[0155] The CPU 606 can execute a sequence of machine-readable instructions that can be embodied in a program or software. The instructions can be stored in a memory location such as the memory 604. The instructions can be directed to the CPU 606 and then program or otherwise configure the CPU 606 to implement the methods of the present disclosure described elsewhere herein. Examples of operations performed by the CPU 606 can include fetch, decode, execute, and write-back.

[0156] The CPU 606 can be part of a circuit such as an integrated circuit. One or more other components of the system 600 can be included in the circuit. In some cases, the circuit is an application-specific integrated circuit (ASIC).

[0157] The storage unit 602 can store files such as drivers, libraries, and saved programs. The storage unit 602 can store user data, for example, user preferences and user programs. In some cases, the computer system 600 can include one or more additional data storage units external to the computer system 600, such as being located on a remote server that communicates with the computer system 600 through an intranet or the Internet.

[0158] The computer system 600 can communicate with one or more remote computer systems via the network 612. For example, the computer system 600 can communicate with a user's remote computer system. Examples of remote computer systems can include a personal computer (e.g., a portable PC), a slate or tablet PC (e.g., an Apple® iPad, a Samsung® Galaxy Tab), a phone, a smartphone (e.g., an Apple® iPhone, an Android-compatible device, a Blackberry®), or a personal digital assistant. A user can access the computer system 600 via the network 612.

[0159] The methods described herein may be implemented by machine (e.g., computer processor) executable code stored on an electronic storage location of a computer system 600, such as, for example, on the memory 604 or the electronic storage unit 602. The machine executable code or machine readable code may be provided in the form of software. In use, the code may be executed by a processor 606. In some cases, the code may be retrieved from the storage unit 602 and stored on the memory 604 for immediate access by the processor 606. In some situations, the electronic storage unit 602 may be excluded and the machine executable instructions may be stored in the memory 604.

[0160] The code may be pre-compiled and configured for use on a machine with a processor adapted to execute the code, or may be compiled during runtime. The code may be provided in a programming language selected to enable the code to be executed in a pre-compiled or as-compiled fashion.

[0161] In some embodiments, the systems described elsewhere herein may include a system for diagnosing one or more target cancerous or non-cancerous health states. In some cases, the system includes (a) one or more processors and (b) non-transitory computer-readable storage medium including software configured to cause the one or more processors to (i) receive one or more nucleic acid molecule sequencing reads of one or more biological samples of the one or more targets, wherein the one or more nucleic acid molecule sequencing reads include sequences of one or more amplified genomic features of one or more non-mammalian nucleic acid molecules, and (ii) output a diagnosis of the cancerous or non-cancerous health state of the one or more targets as a result of providing at least the one or more genomic features of the one or more non-mammalian nucleic acid sequencing reads as an input to a trained prediction model.

[0162] Aspects of the systems and methods provided herein, such as computer system 600, may be embodied in programming. Various aspects of the technology may typically be considered as a "product" or "manufactured article" in the form of machine (or processor) executable code and / or associated data carried or embodied in a type of machine-readable medium. The machine executable code may be stored in an electronic memory unit such as a memory (e.g., read-only memory, random access memory, flash memory) or a hard disk. A "memory" type of medium may include any or all of the tangible memories of a computer, processor, etc., or their associated modules such as various semiconductor memories, tape drives, disk drives, etc., and may provide non-transitory storage at any time for software programming. All or part of the software may sometimes be communicated via the Internet or various other communication networks. Such communication may, for example, enable the loading of software from one computer or processor to another, such as from a management server or host computer to an application server's computer platform. Thus, another type of medium that may carry software elements is used over physical interfaces between local devices, over wired and optical terrestrial communication networks, and over various air links, including light, electrical, and electromagnetic waves. Physical elements that carry such waves, such as wired or wireless links, optical links, etc., may also be considered media that carry software. As used herein, unless limited to non-transitory tangible "memory" media, terms such as "computer" or "machine readable medium" refer to any medium involved in providing instructions to a processor for execution.

[0163] Thus, machine-readable media such as computer-executable code can take many forms including, but not limited to, tangible storage media, carrier wave media, or physical transmission media. Non-volatile storage media can include, for example, optical disks or magnetic disks such as any of the storage devices of any computer(s) that can be used to implement, for example, a database shown in the drawings. Volatile storage media can include dynamic memory such as the main memory of such a computer platform. Tangible transmission media can include coaxial cables, copper wire, and fiber optics, including the wires that comprise a bus within a computer system. Carrier wave transmission media can take the form of electrical or electromagnetic signals, or acoustic or light waves such as those generated during radio frequency (RF) and infrared (IR) data communications. Thus, common forms of computer-readable media include, for example, floppy disks, flexible disks, hard disks, magnetic tape, any other magnetic media, CD-ROM, DVD or DVD-ROM, any other optical media, punch cards, paper tapes, any other physical storage media with patterns of holes, RAM, ROM, PROM, and EPROM, FLASH-EPROM, any other memory chip or cartridge, carrier waves transporting data or instructions, cables or links transporting such carrier waves, or any other media from which a computer can read programming code or data. Many of these forms of computer-readable media can be involved in carrying one or more sequences of one or more instructions to a processor for execution.

[0164] The computer system 600 can include, for example, an electronic display 616 that includes a user interface (UI) 614 for providing, for example, a display for visualizing prediction results or an interface for training a prediction model, or can communicate with the electronic display 616. Examples of UIs include, but are not limited to, graphical user interfaces (GUIs) and web-based user interfaces.

[0165] Preferred embodiments of the present invention are shown and described herein, but it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Here, those skilled in the art will envision numerous variations, modifications, and substitutions without departing from the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be used in practicing the present invention. The following claims define the scope of the present invention, and it is intended that methods and structures within the scope of these claims and their equivalents be covered thereby.

[0166] The methods disclosed herein describe and / or illustrate methods having steps in a finite or specific order, but those skilled in the art will recognize many variations based on the teachings described herein. The steps may be completed in a different order. Steps may be added or deleted. Some of the steps may include sub-steps. Many of the steps may be repeated as frequently as beneficial to the platform.

[0167] One or more of each step of the method or series of operations may be implemented in one or more processors or logic circuits such as programmable array logic for a field programmable gate array, a circuit described herein. The circuit may be programmed to provide one or more of each step of the method or series of operations, and the program may include, for example, program instructions stored in a computer-readable memory, or programmed steps of a logic circuit such as programmable array logic or a field programmable gate array.

[0168] The present invention herein has been described with reference to various exemplary embodiments, but it should be understood that these embodiments are merely illustrative of the principles and applications of the present invention. Those skilled in the art will recognize that various modifications may be made to the exemplary embodiments without departing from the scope of the present invention.

[0169] Furthermore, it should be understood that the various features and / or characteristics of the different embodiments herein can be combined with each other. Thus, it is to be understood that numerous modifications can be made to the exemplary embodiments without departing from the scope of the present invention, and that other configurations can be devised.

[0170] Furthermore, other embodiments of the present invention will be apparent to those skilled in the art from consideration of the specification and embodiments disclosed herein. The specification and examples are considered as illustrative only, and the scope and spirit are intended to be indicated by the following claims.

Example

[0171] Example 1: Enrichment and amplification of microbial nucleic acids result in microbial features that improve predictive model performance Biological samples (e.g., liquid biological samples such as blood) are obtained from subjects characterized and / or classified as being the subject of various clinical classifications, i.e., healthy, having cancer, or having a non-cancerous disease. The biological samples are divided into two groups. The first group consists of biological samples from all healthy classification subjects depleted of human or mammalian DNA bound to nucleosomes. The first group of biological samples is also subjected to further microbial nucleic acid molecule enrichment by use of hybridization probes, protein probes, and / or marker gene amplification described elsewhere herein. The second group consists of biological samples from all healthy classification subjects in which human or mammalian nucleic acid molecules bound to nucleosomes are not depleted and which have not been subjected to further enrichment and / or amplification. The microbial classification and functional characteristics of each group are determined and utilized in combination with the health classification(s) as a label to train two prediction models for classifying the health classification based on the input of microbial classification and functional characteristics. Each prediction model is then tested with a set of subject data and known health classifications to determine the accuracy of the model in classifying the health classification of a subject from the microbial classification and functional characteristics of the subject's biological sample determined therefrom. Prediction models trained with microbial classification and functional characteristics determined from depleted, enriched, and amplified microbial nucleic acids of biological samples are superior to models trained with microbial classification and functional characteristics determined from biological samples not subjected to depletion, enrichment, and / or amplification.

[0172] The increased performance between the two models can be understood as an optimization of the microbial classification and functional characteristics determined from the microbial nucleic acid molecules of the biological samples. By depleting the biological samples of human and / or mammalian nucleic acid molecules, a greater proportion of the sequencing reads remaining in the biological samples will be derived from the microbial content that would otherwise be washed away by the large human host nucleic acid molecule background in each biological sample. By enabling the detection of lower levels of microbial nucleic acid molecules with the methods and systems described herein, subtle differences in microbial classification and / or microbial functional characteristics that more readily distinguish between the microbial composition of a subject and their corresponding health states can be revealed.

[0173] Definition Unless otherwise defined, all technical terms, notations, and other technical, scientific, or specialized terms used in this specification are intended to have the same meaning as commonly understood by one of ordinary skill in the art to which the claimed subject matter pertains. In some cases, terms with commonly understood meanings are defined in this specification for clarity and / or ease of reference, and the inclusion of such definitions in this specification should not necessarily be construed as representing a substantial difference from what is commonly understood in the art.

[0174] Throughout this application, various embodiments may be presented in a range format. It should be understood that the description in range format is merely for convenience and brevity and should not be construed as an inflexible limitation on the scope of the present disclosure. Accordingly, a recitation of a range should be considered to specifically disclose all the possible sub-ranges as well as the individual values within that range. For example, a recitation of a range such as 1 - 6 should be considered to specifically disclose sub-ranges such as 1 - 3, 1 - 4, 1 - 5, 2 - 4, 2 - 6, 3 - 6, as well as the individual numbers within that range, for example, 1, 2, 3, 4, 5, and 6. This applies regardless of the width of the range.

[0175] As used in this specification and the claims, the singular forms "a", "an", and "the" include plural referents unless the context clearly dictates otherwise. For example, the term "a sample" includes a plurality of samples, including mixtures thereof.

[0176] The terms "determining," "measuring," "evaluating," "assessing," "assaying," and "analyzing" are often used interchangeably herein to refer to forms of measurement. These terms include determining whether an element is present (e.g., detecting). These terms can include quantitative, qualitative, or both quantitative and qualitative determinations. Evaluating can be relative or absolute. "Detecting the presence of" can, depending on the context, include determining the amount of what is present in addition to determining whether it is present or not.

[0177] The terms "subject," "individual," or "patient" are used interchangeably herein. A "subject" can be a biological entity that contains expressed genetic material. The biological entity can be a plant, animal, or microorganism, including, for example, bacteria, viruses, fungi, and protozoa. A subject can be a tissue, cell, and their progeny of a biological entity obtained in vivo or cultured in vitro. A subject can be a mammal. The mammal can be a human. A subject can be diagnosed or suspected of being at high risk of a disease. In some cases, a subject is not necessarily diagnosed or suspected of being at high risk of a disease.

[0178] The term "in vivo" describes events occurring within the body of a subject.

[0179] The term "ex vivo" describes events occurring outside the body of a subject. An ex vivo assay is not performed on a subject. Rather, it is performed on a sample separate from the subject. An example of an ex vivo assay performed on a sample is an "in vitro" assay.

[0180] The term "in vitro" is used to describe events that occur within a container that holds laboratory reagents so that the materials are separated from the biological sources from which they are obtained. In vitro assays can include cell-based assays in which live or dead cells are used. In vitro assays can also include cell-free assays in which intact cells are not used.

[0181] As used herein, the term "about" in reference to a number refers to that number plus or minus 10% of that number. The term "about" in reference to a range refers to that range minus 10% of the minimum value thereof and plus 10% of the maximum value thereof.

[0182] As used herein, the terms "treatment" or "treating" are used in reference to a pharmaceutical or other intervention regimen that provides a beneficial or desired result in a recipient. Beneficial or desired results include, but are not limited to, therapeutic and / or prophylactic benefits. A therapeutic benefit may refer to the eradication or amelioration of the treated symptoms or underlying disorder. Additionally, a therapeutic benefit may be achieved by the eradication or amelioration of one or more of the physiological symptoms associated with an underlying disorder such that the subject may still be afflicted with the underlying disorder but improvement is observed in the subject. Prophylactic effects include delaying, preventing, or eliminating the occurrence of a disease or condition, delaying or eliminating the onset of symptoms of a disease or condition, slowing, halting, or reversing the progression of a disease or condition, or any combination thereof. For prophylactic benefits, a subject at risk of developing a particular disease, or a subject reporting one or more of the physiological symptoms of a disease, may be treated even in the absence of a diagnosis of that disease.

[0183] The section headings used herein are for organizational purposes only and are not to be construed as limiting the described subject matter.

[0184] Preferred embodiments of the present invention are shown and described herein, but it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Here, those skilled in the art will envision numerous variations, modifications, and substitutions without departing from the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be used in practicing the present invention. The following claims define the scope of the present invention, and it is intended that methods and structures within the scope of these claims and their equivalents be covered thereby.

Claims

1. A method for generating a microbial metagenomic feature set for diagnosing cancer, wherein the method is (a) To provide the health status of multiple subjects and biological samples taken from said multiple subjects, wherein the biological samples include mammalian nucleic acid molecules and microbial nucleic acid molecules. (b) Removing the mammalian nucleic acid molecules from the biological sample using an affinity capture reagent, (c) Sequencing the microbial nucleic acid molecule to generate a microbial sequencing read, (d) A method comprising generating the set of microbial metagenomic features for diagnosing cancer by combining the abundance of metagenomic features of the microbial sequencing reads with the health status of the plurality of subjects using a computer.

2. The method according to claim 1, wherein the microbial metagenomic feature set includes microbial taxonomic abundances.

3. The method according to claim 1, wherein the set of microbial metagenomic features includes computationally inferred microbial biochemical pathways and associated abundances of the microbial biochemical pathways.

4. The method according to claim 1, wherein the microbial metagenomic feature set includes a microbial phylogenetic marker gene or a fragment of the marker gene.

5. The said biological sample includes a liquid biological sample, and the said liquid biological sample includes plasma, serum, whole blood, urine, cerebrospinal fluid, saliva, sweat, tears, exhaled breath condensate, or any combination thereof, dilutions, or processed fractions thereof. (b) is, (a) The liquid biological sample is brought into contact with a solid support containing an immobilized anti-nucleosome antibody to form an antibody-nucleosome interaction complex, (b) Separating the solid support from the liquid biological sample and concentrating the antibody-nucleosome interaction complex, (c) Purifying one or more remaining nucleosome-depleted microbial nucleic acid molecules, The anti-nucleosome antibody is configured to bind to an epitope containing DNA and one or more histone proteins. The method according to claim 1, wherein the solid support comprises magnetic beads, agarose beads, non-magnetic latex, functionalized Sepharose, pH-sensitive polymer, or any combination thereof.

6. (b) is, (a) Contacting the liquid biological sample with one or more anti-nucleosome antibodies to form an antibody-nucleosome interaction complex, (b) Contacting the antibody-nucleosome interaction complex with a solid support, wherein the surface of the solid support includes a binding portion configured to couple with the antibody-nucleosome interaction complex, (c) Separating the solid support from the liquid biological sample and concentrating the antibody-nucleosome interaction complex, (d) Purifying the remaining nucleosome-depleted microbial nucleic acid molecules, The anti-nucleosome antibody comprises multiple epitope tags, and The plurality of epitope tags include an N-terminal or C-terminal 6×histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), Fc fusion, biotin, or any combination thereof. The solid support comprises magnetic beads, agarose beads, non-magnetic latex, functionalized Sepharose, pH-sensitive polymer, or any combination thereof. The solid support contains an affinity agent immobilized by covalent bonds, or The affinity capture reagent includes an antibody specific to streptavidin, 6× histidine tag, green fluorescent protein (GFP), myc, hemagglutinin (HA), biotin, or any combination thereof. The method according to claim 5, wherein the affinity agent comprises an anti-species antibody.

7. (c) is, (a) Generating a single-stranded DNA library from the microbial nucleic acid molecules, (b) Perform shotgun metagenomic sequencing analysis on the single-stranded DNA library to produce sequencing reads, (c) Filtering the sequencing reads to produce sequencing reads for mammalian DNA-depleted microorganisms, (d) Decontaminating the mammalian DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads, and The aforementioned purification includes in silico purification, or The method according to claim 1, wherein the filtering includes computationally mapping the sequencing reads to a human reference genome database using a computer.

8. (c) is, (a) Amplifying the genomic features of the microbial nucleic acid molecule and thereby generating the amplified genomic features, (b) Sequencing the amplified genome features to generate sequencing reads, (c) Filtering the sequencing reads to produce sequencing reads for mitochondrial DNA-depleted microorganisms, (d) Purifying the mitochondrial DNA-depleted microbial sequencing reads to remove non-endogenous microbial sequencing reads, and The aforementioned purification includes in silico purification, or The aforementioned genome features include a microbial phylogenetic marker gene or a fragment of the marker gene, and The microbial phylogenetic marker gene includes a bacterial marker gene or a fragment of the marker gene, or The microbial phylogenetic marker gene includes a fungal marker gene or a fragment of the marker gene, and The bacterial marker genes include ribosomal RNA gene 5S, ribosomal RNA gene 16S, ribosomal RNA gene 23S, bacterial housekeeping genes dnaG, frr, infC, nusA, pgk, pyrG, rplA, rplB, rplC, rplD, rplE, rplF, rplK, rplL, rplM, rplN, rplP, rplS, rplT, rpmA, rpoB, rpsB, rpsC, rpsE, rpsI, rpsJ, rpsK, rpsM, rpsS, smpB, tsf, or any combination thereof, or The fungal marker gene includes one or more of the ribosomal RNA gene 18S, ribosomal RNA gene 5.8S, ribosomal RNA gene 28S, and internally transcribed spacer regions 1 and 2, or The microbial phylogenetic marker gene includes a marker gene for bacteria, fungi, or any combination thereof, or Amplification includes carrying out a polymerase chain reaction or its derivatives, and The method according to claim 1, wherein the derivative thereof includes reverse PCR, anchored PCR, primer-directed rolling circle amplification, or any combination thereof.

9. (c) includes enriching the microbial nucleic acid molecules, and The aforementioned enrichment, (a) Combining purified nucleosome-depleted microbial nucleic acid molecules with a hybridization probe, wherein the hybridization probe includes nucleic acid sequence complementarity to microbial genome features, (b) Incubating the hybridization probe and the nucleosome-depleted microbial nucleic acid molecule under conditions that promote nucleic acid base pairing between the target nucleic acid features and the hybridization probe, (c) Separating the unbound hybridization probe from the hybridized probe bound to the microbial nucleic acid molecule, (d) Washing the hybridized probe bound to the microbial nucleic acid molecule, thereby generating an enriched microbial nucleic acid molecule, and The washing is to remove nonspecifically associated nucleic acid molecules and other reaction components, or The aforementioned enrichment, (a) Combining purified nucleosome-depleted microbial nucleic acid molecules with recombinant CXXC domain proteins to form a protein-DNA binding reaction, (b) Incubating the protein-DNA binding reaction under conditions that promote interaction between the recombinant CXXC domain protein and the unmethylated CpG motif of the nucleosome-depleted microbial nucleic acid molecule, (c) Separating the unbound recombinant CXXC domain protein and the recombinant CXXC domain protein bound to the unmethylated CpG motif from the remainder of the protein-DNA binding reaction, (d) Washing the recombinant CXXC domain protein bound to the unmethylated CpG motif, thereby generating an enriched nucleic acid molecule for amplification, and The washing is configured to remove the remaining nonspecifically associated nucleic acid molecules and protein-DNA binding reaction components, and The amplification includes carrying out a polymerase chain reaction or its derivatives, and The method according to claim 1, wherein the derivative thereof includes reverse PCR, anchored PCR, primer-directed rolling circle amplification, or any combination thereof.

10. The method according to claim 1, wherein the cancer includes stage I, II, or III cancer.

11. A computer generates a trained predictive model, the trained predictive model being trained on one or more target microbial metagenomic feature sets and health conditions. The aforementioned trained predictive model includes a machine learning model, one or more machine learning models, an ensemble of machine learning models, or any combination thereof, The aforementioned trained predictive model includes a regularized machine learning model, and The machine learning model includes a machine learning classifier, or The method according to claim 1, wherein the machine learning model includes a machine learning model such as a gradient boosting machine, a neural network, a support vector machine, a k-means, a classification tree, a random forest, regression, or any combination thereof.

12. A method for diagnosing a target cancer, wherein the method is (a) To provide a biological sample taken from the subject, wherein the biological sample includes mammalian nucleic acid molecules and microbial nucleic acid molecules. (b) Removing the mammalian nucleic acid molecules from the biological sample using an affinity capture reagent, (c) Sequencing multiple microbial nucleic acid molecules of the biological sample to generate microbial sequencing reads, (d) To generate metagenomic feature abundances of the microbial sequencing reads, (e) A method comprising outputting the diagnosis of the cancer in question as a result of providing the metagenomic feature abundance as input to a predictive model trained by computer execution.

13. A system for diagnosing a target cancer, wherein the system (a) Processor and (b) The processor: (i) Receiving a target mammalian nucleosome-depleted nucleic acid molecule sequencing read, wherein the mammalian nucleosome-depleted nucleic acid molecule sequencing read includes metagenomic features of a microbial nucleic acid molecule, and (ii) A system including a non-temporary computer-readable storage medium, which includes software configured to output a diagnosis of the cancer in question as a result of providing the metagenomic features as input to a trained predictive model.

14. A method for generating metagenomic features of a sample of cell-free microbial nucleic acid molecules for diagnosing non-tumor diseases, (a) Contacting a sample of cell-free microbial nucleic acid molecules with a probe, wherein the probe includes a binding portion configured to bind to human nucleic acid molecules complexed with a protein, (b) Remove the probe bound to the human nucleic acid molecule complexed with the protein, thereby producing an enriched cell-free microbial nucleic acid molecule, (c) A method comprising generating metagenomic features of the enriched cell-free microbial nucleic acid molecules configured by computer for diagnosing non-neoplastic diseases.

15. A method for generating a microbial metagenomic feature set for diagnosing non-tumor diseases, wherein the method is (a) To provide the health status of multiple subjects and biological samples taken from said multiple subjects, wherein the biological samples include mammalian nucleic acid molecules and microbial nucleic acid molecules. (b) Removing the mammalian nucleic acid molecules from the biological sample using an affinity capture reagent, (c) Sequencing the microbial nucleic acid molecule to generate a microbial sequencing read, (d) A method comprising generating a set of microbial metagenomic features for diagnosing a non-neoplastic disease by combining the abundance of metagenomic features of the microbial sequencing reads with the health status of the plurality of subjects using a computer.

Images