Multi-omics evaluation

Abstract

Methods such as multi-omics methods for evaluating diseases are described herein. In multi-omics methods, proteomics data, transcriptomics data, genomics data, lipidomics data, or metabolomics data can be integrated. A method for screening a disease or a disease state. Also described herein is a method for screening a disease or a disease state from a biological sample. Also described herein are multi-omics databases and methods of using them.

Description

【Technical Field】 【0001】 Cross-reference This application claims the benefit of U.S. Provisional Application No. 63 / 349,937, filed Jun. 7, 2022; U.S. Provisional Application No. 63 / 399,998, filed Aug. 22, 2022; and U.S. Provisional Application No. 63 / 486,247, filed Feb. 21, 2023, which are hereby incorporated by reference in their entirety. 【0002】 Incorporation of Sequence Listing by Reference This application has been filed with an electronic sequence listing. The sequence listing is provided as a file named "PrognomIQ 712.601 sequence listing.xml", created on Jun. 3, 2023, and is 160,536 bytes in size. The information in the electronic sequence listing is hereby incorporated by reference in its entirety. 【Background Art】 【0003】 Background There is a need for methods to accurately detect disease states, such as cancer, at an early stage. Accurate and early disease detection can improve the treatment and prognosis of subjects with the disease. 【Summary of the Invention】 【Means for Solving the Problems】 【0004】 Summary In some aspects, the present specification discloses a multi-omics method. This method can be useful for biomarker discovery or for assessing a disease or a disease state. Some aspects include obtaining a multi-omics database comprising multi-omics data generated from biological fluid samples of a population having various disease states and patient characteristics, and querying the multi-omics database to identify a biomarker or a set of biomarkers capable of distinguishing individuals of a population having a first disease state or patient characteristic from other individuals of a population having a second disease state or patient characteristic. In some aspects, the multi-omics data includes proteomics, metabolomics, lipidomics, transcriptomics, fragmentomics, methylomics, or genomics, or combinations thereof. The multi-omics data includes proteomics, metabolomics, lipidomics, transcriptomics, fragmentomics, methylomics, and genomics. In some aspects, the step of querying includes identifying a biomarker or a set of biomarkers as useful for identifying a third disease state or patient characteristic, and determining that the biomarker or the set of biomarkers is also useful for identifying the first or second first disease state or patient characteristic. In some aspects, the step of querying includes identifying another biomarker or a set of biomarkers as useful for distinguishing individuals of a population having a first disease state or patient characteristic from other individuals of a population having a second disease state or patient characteristic, and determining that the biomarker or the set of biomarkers correlates with another biomarker or a set of biomarkers among the individuals of the population. In some aspects, the step of querying includes comparing or correlating measurements of the multi-omics data. In some aspects, the step of querying the multi-omics database includes correlating the values of the multi-omics data with the first or second disease state or patient characteristic. In some aspects, the step of querying includes the use of machine learning.In some embodiments, the multi-omics data is generated from biological fluid samples of more than 500, more than 1000, more than 1500, more than 2000, more than 2500, or more than 3000 members of a population. In some embodiments, the multi-omics data is generated from biological fluid samples of 500 or fewer, 1000 or fewer, 1500 or fewer, 2000 or fewer, 2500 or fewer, or 3000 or fewer members of a population. In some embodiments, the multi-omics data is generated using a non-targeted omics measurement method. In some embodiments, at least some of the multi-omics data is generated after using nanoparticle enrichment. In some embodiments, a biomarker or a set of biomarkers includes secreted biomarkers. In some embodiments, a biomarker or a set of biomarkers includes proteins, lipids, nucleic acids, metabolites, or combinations thereof. In some embodiments, a set of biomarkers corresponds to a metabolic pathway. In some embodiments, the first disease state or patient characteristic includes a cancer state. In some embodiments, the first or second disease state or patient characteristic includes a co-existing condition. In some embodiments, the second disease state or patient characteristic includes a healthy state. In some embodiments, the first or second patient characteristic includes age, gender, race, weight, height, diet intake, exercise habits, activity level, or smoking status. Some embodiments further include the step of using a biomarker or a set of biomarkers to classify a subject as having a first disease state or patient characteristic or as having a second disease state or patient characteristic. Some embodiments further include the step of identifying, recommending, or administering a disease treatment based on the use of a biomarker or a set of biomarkers. In some embodiments, the biological fluid sample includes a blood, serum, or plasma sample. In some embodiments, the population includes human subjects. 【0005】 In some embodiments, the present specification discloses a method that includes the steps of obtaining multi-omics data from one or more biofluid samples of a subject identified as having a pulmonary nodule, and applying a classifier to the multi-omics data to evaluate whether the pulmonary nodule is cancerous or non-cancerous. In some embodiments, the multi-omics data includes metabolomics data, lipidomics data, proteomics data, or transcriptomics data. In some embodiments, the proteomics data includes targeted proteomics data. In some embodiments, the proteomics data includes untargeted proteomics data. In some embodiments, the transcriptomics data includes mRNA data. In some embodiments, the transcriptomics data includes microRNA data. In some embodiments, the classifier, when distinguishing whether a biofluid sample indicates a cancerous pulmonary nodule, is at least about 0 when determined in the receiver operating characteristic curve.It functions with the area under the curve of 6. Any biomarker or biomarkers disclosed in this specification can be used in the evaluation or as a feature of the classifier feature. Some examples of biomarkers that can be included in multi-omics data are STVLTIPEIIIK (SEQ ID NO: 12), TLAFPLTIR (SEQ ID NO: 13), LIQGAPTIR (SEQ ID NO: 14), SSGLVSNAPGVQIR (SEQ ID NO: 15), DGSFSVVITGLR (SEQ ID NO: 16), LGPISADSTTAPLEK (SEQ ID NO: 17), SEAACLAAGPGIR (SEQ ID NO: 18), TDTGFLQTLGHNLFGIYQK (SEQ ID NO: 19), LKPEDITQIQPQQLVLR (SEQ ID NO: 20), GLPAPIEK (SEQ ID NO: 21), LLGPGPAADFSVSVER (SEQ ID NO: 22), YEYLEGGDR (SEQ ID NO: 23), HLEDVFSK (SEQ ID NO: 24), ILGPLSYSK (SEQ ID NO: 25), NCQTVLAPCSPNPCENAAVCK (SEQ ID NO: 26), TVTATFGYPFR (SEQ ID NO: 27), STDTSCVNPPTVQNAHILSR (SEQ ID NO: 28), FSLVSGWGQLLDR (SEQ ID NO: 29), ELLALIQLER (SEQ ID NO: 30), DAHSVLLSHIFHGR (SEQ ID NO: 31), EHAVEGDCDFQLLK (SEQ ID NO: 32), SQASSCSLQSSDSVPVGLCK (SEQ ID NO: 33), GEFAIDGYSVR (SEQ ID NO: 34), ALVEGVDQLFTDYQIK (SEQ ID NO: 35), LLPYIVGVAQR (SEQ ID NO: 36), HTLNQIDEVK (SEQ ID NO: 37), IDILVNNGGMSQR (SEQ ID NO: 38), LMMDGHEVTVVDNFFTGR (SEQ ID NO: 39), MYGEILSPNYPQAYPSEVEK (SEQ ID NO: 40), NNEEWTVDSCTECHCQNSVTICK (SEQ ID NO: 41), IDTQDIEASHYR (SEQ ID NO: 42), TFIFSDLDYMGMSSGFYK (SEQ ID NO: 43), PDAELSASSVYNLLPEK (SEQ ID NO: 44), ASIHEAWTDGK (SEQ ID NO: 45), LYPWGVVEVENPEHNDFLK (SEQ ID NO: 46), YHWEHTGLTLR (SEQ ID NO: 47), or IGGAIEEVYVSLGVSVGK (SEQ ID NO: 48), ENSG00000224067.2. It may include the measured values of biomarkers selected from the group consisting of ENSG00000196735.13, ENSG00000287647.1, ENSG00000230797.3, ENSG00000287219.1, ENSG00000271543.1, ENSG00000223711.1, ENSG00000177602.5, ENSG00000144671.11, ENSG00000129673.10, ENSG00000265817.4, ENSG00000108924.14, ENSG00000232125.5, ENSG00000252800.1, ENSG00000287537.1, ENSG00000196405.13, ENSG00000250893.1, ENSG00000153446.16, ENSG00000284630.1, ENSG00000284687.1, PC(20:3_20:4)+AcO, DAG(18:2_20:2)+NH4, PC(18:2_20:5)+AcO, LPE(18:1)-H, LPE(16:0)-H, TAG(58:6_FA18:0)+NH4, DAG(20:1_20:5)+NH4, PC(14:0_20:2)+AcO, PC(18:2_20:3)+AcO, PE(18:1_22:4)-H, PE(18:0_20:1)-H, CER(d18:1 / 26:1)+H, PC(14:0_18:2)+AcO, PE(18:0_22:4)-H, PI(15:0_22:5)-H, PE(P-18:1_18:0)+H, TAG(54:5_FA18:3)+NH4, TAG(58:5_FA18:1)+NH4, DAG(20:5_22:4)+NH4, LPE(20:3)-H. The biomarker may include PC(20:3_20:4)+AcO, sedoheptulose 1,7-bisphosphate, glucuronate, biopterin, reduced glutathione, N-acetyl-arginine, cotinine, indole-3-lactate, 13C4-oxoglutarate, propionyl-CoA, AICAR, 3-methyl-3-hydroxyglutaric acid, imidazoleacetic acid, shikimic acid, 1-methyladenosine, dopamine, carnosine, homocitrulline, indolepyruvate, 2-phosphoglycerate, or glutaric acid. 【0006】 In some aspects, the present specification discloses a method including the steps of obtaining multi-omics data from one or more biological fluid samples of a subject suspected of having pancreatic cancer, and applying a classifier to the multi-omics data to evaluate the likelihood that the subject has pancreatic cancer. In some aspects, when determining whether a biological fluid sample indicates pancreatic cancer, the classifier has a receiver operating characteristic curve of at least 0.85, at least 0.86, at least 0.87, at least 0.88, at least 0.89, at least 0.90, at least 0.91, at least 0.92, at least 0.93, at least 0.94, at least 0.95, at least 0.96, at least 0.97, or at least 0.It functions with the area under the curve of 98. In some embodiments, pancreatic cancer includes pancreatic cancer at stage 1 or 2. In some embodiments, pancreatic cancer includes pancreatic cancer at stage 3 or 4. In some embodiments, the multi-omics data includes data related to copy number polymorphisms, fragmentomics, mRNA, proteins, metabolites, or lipids. In some embodiments, the multi-omics data includes copy number polymorphism data, fragmentomics data, transcriptomics data, proteomics data, metabolic data, and lipidomics data. Any biomarker or biomarkers disclosed herein can be used as a feature in an assessment or in a classifier feature. Some examples of biomarkers that can be included in the multi-omics data are P00488, P15144, P01833, P58335, P05109, P02750, O95445, P02654, P06702, O14786, P08637, P02766, Q9NQ79, P05362, Q13740, P24821, P06396, P05452, P18065, Q8WWA0, Q06033, P19320, P02656, Q01628, P01011, Q9H4F8, P01009, P26022, Q9BYE9, Q16777, P09237, P10643, P07355, Q08830, P62805, P49748, TELVEPTEYLVVHLK (SEQ ID NO: 1), TFVIIPELVLPNR (SEQ ID NO: 2), LQELHLSSNGLESLSPEFLRPVPQLR (SEQ ID NO: 3), ITLLSALVETR (SEQ ID NO: 4), VVATTQMQAADAR (SEQ ID NO: 5), TFVIIPELVLPNR (SEQ ID NO: 6), LQHLENELTHDIITK (SEQ ID NO: 7), FLENEDRR (SEQ ID NO: 8), LWYENPGVFSPAQLTQIK (SEQ ID NO: 9), QWMENPNNNPIHPNLR (SEQ ID NO: 10), LEIYQEDQIHFMCPLAR (SEQ ID NO: 11), ENSG00000170088.14, ENSG00000274641.2, ENSG00000248180.1, ENSG00000271270.7, ENSG00000132846.6, ENSG00000280247.1, ENSG00000284035.1, ENSG00000277681.1, ENSG00000264559.1, ENSG00000264764.1, ENSG00000216101.3, ENSG00000266297.1, ENSG00000266320.1, ENSG00000273836.1, ENSG00000199135.1, ENSG00000207604.3, ENSG00000221656.1, ENSG00000207639.1, ENSG00000207607.3, ENSG00000199121.4, ENSG00000265253.1, ENSG00000283728.1, ENSG00000207563.1, ENSG00000283978.1, ENSG00000208015.1, ENSG00000207993.3, ENSG00000208012.1, ENSG00000284195.1, ENSG00000264796.1, ENSG00000278549.1, ENSG00000283764.1, ENSG00000221540.1, ENSG00000263381.1, ENSG00000265435.1, ENSG00000198976.1, ENSG00000208037.1, ENSG00000207757.1, ENSG00000263409.1, ENSG00000221493.1, ENSG00000207807.1, ENSG00000207870.1, CER(d18:1 / 16:0)+H, CER(d18:1 / 16:0)+H, CER(d18:1 / 18:0)+H, PA(18:0_20:5)-H, DAG(18:1_20:0)+NH4, PC(18:2_20:5)+AcO, PC(20:3_20:4)+AcO, PE(O-18:0_22:5)-H, PE(14:0_22:5)-H, PC(16:0_20:2)+AcO, PI(18:3+20:4)-H, PA(20:2+20:3)-H, 17:0-18:1PE-d5-H_USPLASH.It may contain IS, PC(16:0_16:0)+AcO, PC(17:0_20:1)+AcO, CER(d18:0 / 24:0)+H, PE(P=16:0+22:5)+H, PE(18:2+20:1)-H, PE(P-16:0+20:5)+H, TAG(48:0+FA16:0)+NH4, PC(16:0+18:1)+AcO, PE(18:0+20:2)=H, PE(18:1+20:1)-H, AICAR, CMP, dimethylglycine, epinephrine, sorbitol, 5-thymidylic acid (dTMP), tauro-muricholic acid, glycocholate, fructose-6-phosphate, farnesyl pyrophosphate, ATP, cystamine, taurocholate, glycine, choline, hydroxyphenyl lactic acid, inosine, glutaryl carnitine, 1-methylimidazole acetic acid, AMP, gluconate, reduced glutathione, glutamic acid, creatine, L-dihydroorotate, thymidine, imidazole acetic acid, or UMP. 【0007】 In some aspects, a method for detecting pancreatic cancer is disclosed herein. The method includes obtaining a biomarker from a biological fluid sample of a subject and applying a classifier to the biomarker, the applying including evaluating pancreatic cancer, the classifier having a performance characterized by a receiver operating characteristic (ROC) curve with an average or median area under the curve (AUC) of at least 0.7 to distinguish between biological fluid samples of subjects with and without pancreatic cancer, the biomarker being any of the following chromosomal regions: ThXX chr10:113000001-113100000, chr7:45200001-45300000, chr9:104900001-105000000, chr18:58600001-58700000, chr17:17400001-17500000, chr2:150700001-150800000, chr7:149300001-149400000, chr4:88700001-88800000, chr20:28900001-29000000, or chr8:55300001-55400000; any of the following mRNA transcripts: TMEM192, H2BC17, GAPDHP60, ENSG00000271270.7, ZBED3, or GRCh38; any of the following microRNAs: MIR5187, MIR6739, MIR3162, MIR4772, MIR877, MIR744, MIR3909, MIR6842, MIR101-1, MIR206, MIR1225, MIR193B, MIR200A, MIR26B, MIR4446, MIR7108, MIR23B, MIR365B, MIR362, MIR134, MIRLET7F2, MIR6852, MIR5009, MIR6736, MIR6850, MIR1180, MIR5584, MIR3121, MIR429, MIR320A, MIR93, MIR4747, MIR320C1, or MIR221; any of the following proteins F13A_HUMAN, AMPN_HUMAN, PIGR_HUMAN, ANTR2_HUMAN, S10A8_HUMAN, A2GL_HUMAN, APOM_HUMAN, APOC1_HUMAN, S10A9_HUMAN, NRP1_HUMAN, FCG3A_HUMAN, TTHY_HUMAN,Any one of CRAC1_HUMAN, ICAM1_HUMAN, CD166_HUMAN, TENA_HUMAN, GELS_HUMAN, TETN_HUMAN, IBP2_HUMAN, ITLN1_HUMAN, ITIH3_HUMAN, VCAM1_HUMAN, or APOC3_HUMAN, any one of the following peptides: TELVEPTEYLVVHLK (SEQ ID NO: 1), TFVIIPELVLPNR (SEQ ID NO: 2), LQELHLSSNGLESLSPEFLRPVPQLR (SEQ ID NO: 3), ITLLSALVETR (SEQ ID NO: 4), VVATTQMQAADAR (SEQ ID NO: 5), TFVIIPELVLPNR (SEQ ID NO: 6), LQHLENELTHDIITK (SEQ ID NO: 7), FLENEDRR (SEQ ID NO: 8), LWYENPGVFSPAQLTQIK (SEQ ID NO: 9), QWMENPNNNPIHPNLR (SEQ ID NO: 10), or LEIYQEDQIHFMCPLAR (SEQ ID NO: 11), any one of the following proteins: IFM3_HUMAN, AMPN_HUMAN, A2GL_HUMAN, AACT_HUMAN, SMOC1_HUMAN, A1AT_HUMAN, PTX3_HUMAN, CDHR2_HUMAN, H2A2C_HUMAN, ANTR2_HUMAN, MMP7_HUMAN, CO7_HUMAN, ANXA2_HUMAN, FGL1_HUMAN, H4_HUMAN, or ACADV_HUMAN, the following lipids: CER(d18:1 / 16:0)+H, CER(d18:1 / 18:0)+H, PA(18:0_20:5)-H, DAG(18:1_20:0)+NH4, PC(18:2_20:5)+AcO, PC(20:3_20:4)+AcO, PE(O-18:0_22:5)-H, PE(14:0_22:5)-H, PC(16:0_20:2)+AcO, PI(18:3+20:4)-H, PA(20:2+20:3)-H, 17:0-18:1 PE-d5-H_USPLASH.IS, PC(16:0_16:0)+AcO, PC(17:0_20:1)+AcO, CER(d18:0 / 24:0)+H, PE(P=16:0+22:5)+H, PE(18:2+20:1)-H, PE(P-16:0+20:5)+H, TAG(48:0+FA16:0)+NH4, PC(16:0+18:1)+AcO,Or any of PE(18:0+20:2)=H, PE(18:1+20:1)-H, or the following metabolites: AICAR, CMP, dimethylglycine, epinephrine, sorbitol, 5-thymidylic acid (dTMP), tauro-muricholic acid, glycocholate, fructose-6-phosphate, farnesyl pyrophosphate, ATP, cystamine, taurocholate, glycine, choline, hydroxyphenyl lactic acid, inosine, glutaryl carnitine, 1-methylimidazole acetic acid, AMP, gluconate, reduced glutathione, glutamic acid, creatine, L-dihydroorotate, thymidine, imidazole acetic acid, or UMP. In some embodiments, the biomarker comprises any of the following chromosomal regions: chr10:113000001~113100000, chr7:45200001~45300000, chr9:104900001~105000000, chr18:58600001~58700000, chr17:17400001~17500000, chr2:150700001~150800000, chr7:149300001~149400000, chr4:88700001~88800000, chr20:28900001~29000000, or chr8:55300001~55400000. In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 chromosomal regions. In some embodiments, the biomarker comprises any of the following mRNA transcripts: TMEM192, H2BC17, GAPDHP60, ENSG00000271270.7, ZBED3, or GRCh38. In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, or 6 mRNA transcripts. In some embodiments, the biomarker comprises the following microRNAs: MIR5187, MIR6739, MIR3162, MIR4772, MIR877, MIR744, MIR3909, MIR6842, MIR101-1, MIR206, MIR1225, MIR193B, MIR200A, MIR26B, MIR4446, MIR7108, MIR23B, MIR365B, MIR362, MIR134, MIRLET7F2, MIR6852, MIR5009, MIR6736, MIR6850,It comprises any one of MIR1180, MIR5584, MIR3121, MIR429, MIR320A, MIR93, MIR4747, MIR320C1, or MIR221. In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, or 33 microRNAs. In some embodiments, the biomarker comprises any one of the following proteins: F13A_HUMAN, AMPN_HUMAN, PIGR_HUMAN, ANTR2_HUMAN, S10A8_HUMAN, A2GL_HUMAN, APOM_HUMAN, APOC1_HUMAN, S10A9_HUMAN, NRP1_HUMAN, FCG3A_HUMAN, TTHY_HUMAN, CRAC1_HUMAN, ICAM1_HUMAN, CD166_HUMAN, TENA_HUMAN, GELS_HUMAN, TETN_HUMAN, IBP2_HUMAN, ITLN1_HUMAN, ITIH3_HUMAN, VCAM1_HUMAN, or APOC3_HUMAN. In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, or 23 proteins. In some embodiments, the biomarker comprises any one of the following peptides: TELVEPTEYLVVHLK (SEQ ID NO: 1), TFVIIPELVLPNR (SEQ ID NO: 2), LQELHLSSNGLESLSPEFLRPVPQLR (SEQ ID NO: 3), ITLLSALVETR (SEQ ID NO: 4), VVATTQMQAADAR (SEQ ID NO: 5), TFVIIPELVLPNR (SEQ ID NO: 6), LQHLENELTHDIITK (SEQ ID NO: 7), FLENEDRR (SEQ ID NO: 8), LWYENPGVFSPAQLTQIK (SEQ ID NO: 9), QWMENPNNNPIHPNLR (SEQ ID NO: 10), or LEIYQEDQIHFMCPLAR (SEQ ID NO: 11). In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or 11 peptides. In some embodiments, the biomarker comprises the following proteins IFM3_HUMAN, AMPN_HUMAN,It comprises any one of A2GL_HUMAN, AACT_HUMAN, SMOC1_HUMAN, A1AT_HUMAN, PTX3_HUMAN, CDHR2_HUMAN, H2A2C_HUMAN, ANTR2_HUMAN, MMP7_HUMAN, CO7_HUMAN, ANXA2_HUMAN, FGL1_HUMAN, H4_HUMAN, or ACADV_HUMAN. In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, or 16 proteins. In some embodiments, the biomarker comprises any one of the following lipids: CER(d18:1 / 16:0)+H, CER(d18:1 / 18:0)+H, PA(18:0_20:5)-H, DAG(18:1_20:0)+NH4, PC(18:2_20:5)+AcO, PC(20:3_20:4)+AcO, PE(O-18:0_22:5)-H, PE(14:0_22:5)-H, PC(16:0_20:2)+AcO, PI(18:3+20:4)-H, PA(20:2+20:3)-H, 17:0-18:1PE-d5-H_USPLASH.IS, PC(16:0_16:0)+AcO, PC(17:0_20:1)+AcO, CER(d18:0 / 24:0)+H, PE(P=16:0+22:5)+H, PE(18:2+20:1)-H, PE(P-16:0+20:5)+H, TAG(48:0+FA16:0)+NH4, PC(16:0+18:1)+AcO, PE(18:0+20:2)=H, or PE(18:1+20:1)-H. In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, or 22 lipids. In some embodiments, the biomarker comprises the following metabolites: AICAR, CMP, dimethylglycine, epinephrine, sorbitol, 5-thymidylic acid (dTMP), tauro-muricholic acid, glycocholate, fructose-6-phosphate, farnesyl pyrophosphate, ATP, cystamine, taurocholate, glycine, choline, hydroxyphenyl lactic acid, inosine, glutaryl carnitine, 1-methylimidazole acetic acid, AMP, gluconate, reduced glutathione, glutamic acid, creatine, L-dihydroorotate, thymidine,It contains either imidazole acetic acid or UMP. In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, or 28 metabolites. In some embodiments, the biomarker is the following biomarker, - marker: any one of APOM_HUMAN, G6PE_HUMAN, F13A_HUMAN, A1AT_HUMAN, AACT_HUMAN, A2MG_HUMAN, CO5_HUMAN, IGHG2_HUMAN, APOC1_HUMAN, APOC3_HUMAN, APOB_HUMAN, ICAM1_HUMAN, ITB1_HUMAN, GELS_HUMAN, S10A9_HUMAN, CO8B_HUMAN, TSP1_HUMAN, MMP7_HUMAN, or CO7_HUMAN. In some embodiments, the classifier comprises a performance characterized by a receiver operating characteristic (ROC) curve having an average or median area under the curve (AUC) of at least 0.8. In some embodiments, the subject is suspected of having pancreatic cancer. In some embodiments, evaluating comprises identifying a biomarker indicative of pancreatic cancer. In some embodiments, the method further comprises treating the subject for pancreatic cancer if the subject has pancreatic cancer. In some embodiments, the method further comprises monitoring the subject if the subject does not have pancreatic cancer. 【0008】 In some aspects, a method for detecting lung cancer is disclosed herein. The method includes the steps of identifying a biomarker from a biological fluid sample of a subject and applying a classifier to the biomarker to evaluate for lung cancer, the classifier having a performance characterized by a receiver operating characteristic (ROC) curve with an average or median area under the curve (AUC) of at least 0.7 to distinguish between biological fluid samples of subjects with and without lung cancer, and the biomarker being any of the following: antisense to RNA BAT2, HLA-DQA1, AK5, YY2, ENSG00000287219.1, RPL6 pseudogene, ENSG00000223711.1, HASPIN, SLC22A14, AANAT, FSBP, HLF, DYTN, ENSG00000252800.1, a novel human transcript derived from positions 49,536,677 to 49,538,894 on the reverse strand of chromosome 12 of genome build GRCh38, EVL, a novel human transcript derived from positions 40,426,119 to 40,427,585 on the forward strand of chromosome 4 of genome build GRCh38, C16orf89, a novel human transcript derived from positions 21,657,811 to 21,661,021 on the forward strand of chromosome 22 of genome build GRCh38, or RBFOX1; any of the following lipids: PC(20:3_20:4)+AcO, DAG(18:2_20:2)+NH4, PC(18:2_20:5)+AcO, LPE(18:1)-H, LPE(16:0)-H, TAG(58:6_FA18:0)+NH4, DAG(20:1_20:5)+NH4, PC(14:0_20:2)+AcO, PC(18:2_20:3)+AcO, PE(18:1_22:4)-H, PE(18:0_20:1)-H, CER(d18:1 / 26:1)+H, PC(14:0_18:2)+AcO, PE(18:0_22:4)-H, PI(15:0_22:5)-H, PE(P-18:1_18:0)+H, TAG(54:5_FA18:3)+NH4, TAG(58:5_FA18:1)+NH4, DAG(20:5_22:4)+NH4, or LPE(20:3)-H; any of the following metabolites: sedoheptulose 1,7-bisphosphate, glucuronate, biopterin, reduced glutathione, N-acetyl-arginine,Any of cotinine, indole-3-lactate, 13C4-oxoglutarate, propionyl-CoA, AICAR, 3-methyl-3-hydroxyglutaric acid, imidazoleacetic acid, shikimic acid, 1-methyladenosine, dopamine, carnosine, homocitrulline, indolepyruvate, 2-phosphoglycerate, or glutaric acid, the following peptides STVLTIPEIIIK (SEQ ID NO: 12), TLAFPLTIR (SEQ ID NO: 13), LIQGAPTIR (SEQ ID NO: 14), SSGLVSNAPGVQIR (SEQ ID NO: 15), DGSFSVVITGLR (SEQ ID NO: 16), LGPISADSTTAPLEK (SEQ ID NO: 17), SEAACLAAGPGIR (SEQ ID NO: 18), TDTGFLQTLGHNLFGIYQK (SEQ ID NO: 19), LKPEDITQIQPQQLVLR (SEQ ID NO: 20), GLPAPIEK (SEQ ID NO: 21), LLGPGPAADFSVSVER (SEQ ID NO: 22), YEYLEGGDR (SEQ ID NO: 23), HLEDVFSK (SEQ ID NO: 24), ILGPLSYSK (SEQ ID NO: 25), NCQTVLAPCSPNPCENAAVCK (SEQ ID NO: 26), TVTATFGYPFR (SEQ ID NO: 27), STDTSCVNPPTVQNAHILSR (SEQ ID NO: 28), FSLVSGWGQLLDR (SEQ ID NO: 29), ELLALIQLER (SEQ ID NO: 30), or DAHSVLLSHIFHGR (SEQ ID NO: 31), or any of their fragments, or the following peptides EHAVEGDCDFQLLK (SEQ ID NO: 32), SQASSCSLQSSDSVPVGLCK (SEQ ID NO: 33), GEFAIDGYSVR (SEQ ID NO: 34), ALVEGVDQLFTDYQIK (SEQ ID NO: 35), LLPYIVGVAQR (SEQ ID NO: 36), HTLNQIDEVK (SEQ ID NO: 37), IDILVNNGGMSQR (SEQ ID NO: 38), LMMDGHEVTVVDNFFTGR (SEQ ID NO: 39), MYGEILSPNYPQAYPSEVEK (SEQ ID NO: 40), NNEEWTVDSCTECHCQNSVTICK (SEQ ID NO: 41), IDTQDIEASHYR (SEQ ID NO: 42), TFIFSDLDYMGMSSGFYK (SEQ ID NO: 43), PDAELSASSVYNLLPEK (SEQ ID NO: 44), ASIHEAWTDGK (SEQ ID NO: 45),A method comprising LYPWGVVEVENPEHNDFLK (SEQ ID NO: 46), YHWEHTGLTLR (SEQ ID NO: 47), or IGGAIEEVYVSLGVSVGK (SEQ ID NO: 48), or any of their fragments. In some embodiments, the biomarker comprises the following RNAs: BAT2, HLA-DQA1, antisense of AK5, YY2, ENSG00000287219.1, RPL6 pseudogene, ENSG00000223711.1, HASPIN, SLC22A14, AANAT, FSBP, HLF, DYTN, ENSG00000252800.1, a novel human transcript derived from positions 49,536,677 to 49,538,894 of the reverse strand of chromosome 12 in genome build GRCh38, EVL, a novel human transcript derived from positions 40,426,119 to 40,427,585 of the forward strand of chromosome 4 in genome build GRCh38, C16orf89, a novel human transcript derived from positions 21,657,811 to 21,661,021 of the forward strand of chromosome 22 in genome build GRCh38, or RBFOX1. In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 RNAs. In some embodiments, the biomarker comprises any of the following lipids: PC(20:3_20:4)+AcO, DAG(18:2_20:2)+NH4, PC(18:2_20:5)+AcO, LPE(18:1)-H, LPE(16:0)-H, TAG(58:6_FA18:0)+NH4, DAG(20:1_20:5)+NH4, PC(14:0_20:2)+AcO, PC(18:2_20:3)+AcO, PE(18:1_22:4)-H, PE(18:0_20:1)-H, CER(d18:1 / 26:1)+H, PC(14:0_18:2)+AcO, PE(18:0_22:4)-H, PI(15:0_22:5)-H, PE(P-18:1_18:0)+H, TAG(54:5_FA18:3)+NH4, TAG(58:5_FA18:1)+NH4, DAG(20:5_22:4)+NH4, or LPE(20:3)-H. In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,It contains 16, 17, 18, 19, or 20 lipids. In some embodiments, the biomarker comprises any one of the following metabolites: sedoheptulose 1,7-bisphosphate, glucuronate, biopterin, reduced glutathione, N-acetyl-arginine, cotinine, indole-3-lactate, 13C4-oxoglutarate, propionyl-CoA, AICAR, 3-methyl-3-hydroxyglutaric acid, imidazoleacetic acid, shikimic acid, 1-methyladenosine, dopamine, carnosine, homocitrulline, indolepyruvate, 2-phosphoglyceric acid, or glutaric acid. In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 metabolites. In some embodiments, the biomarker comprises any one of the following peptides: STVLTIPEIIIK (SEQ ID NO: 12), TLAFPLTIR (SEQ ID NO: 13), LIQGAPTIR (SEQ ID NO: 14), SSGLVSNAPGVQIR (SEQ ID NO: 15), DGSFSVVITGLR (SEQ ID NO: 16), LGPISADSTTAPLEK (SEQ ID NO: 17), SEAACLAAGPGIR (SEQ ID NO: 18), TDTGFLQTLGHNLFGIYQK (SEQ ID NO: 19), LKPEDITQIQPQQLVLR (SEQ ID NO: 20), GLPAPIEK (SEQ ID NO: 21), LLGPGPAADFSVSVER (SEQ ID NO: 22), YEYLEGGDR (SEQ ID NO: 23), HLEDVFSK (SEQ ID NO: 24), ILGPLSYSK (SEQ ID NO: 25), NCQTVLAPCSPNPCENAAVCK (SEQ ID NO: 26), TVTATFGYPFR (SEQ ID NO: 27), STDTSCVNPPTVQNAHILSR (SEQ ID NO: 28), FSLVSGWGQLLDR (SEQ ID NO: 29), ELLALIQLER (SEQ ID NO: 30), or DAHSVLLSHIFHGR (SEQ ID NO: 31), or a fragment thereof. In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 peptides. In some embodiments, the biomarker comprises the following peptide: EHAVEGDCDFQLLK (SEQ ID NO: 32),SQASSCSLQSSDSVPVGLCK (SEQ ID NO: 33), GEFAIDGYSVR (SEQ ID NO: 34), ALVEGVDQLFTDYQIK (SEQ ID NO: 35), LLPYIVGVAQR (SEQ ID NO: 36), HTLNQIDEVK (SEQ ID NO: 37), IDILVNNGGMSQR (SEQ ID NO: 38), LMMDGHEVTVVDNFFTGR (SEQ ID NO: 39), MYGEILSPNYPQAYPSEVEK (SEQ ID NO: 40), NNEEWTVDSCTECHCQNSVTICK (SEQ ID NO: 41), IDTQDIEASHYR (SEQ ID NO: 42), TFIFSDLDYMGMSSGFYK (SEQ ID NO: 43), PDAELSASSVYNLLPEK (SEQ ID NO: 44), ASIHEAWTDGK (SEQ ID NO: 45), LYPWGVVEVENPEHNDFLK (SEQ ID NO: 46), YHWEHTGLTLR (SEQ ID NO: 47), or IGGAIEEVYVSLGVSVGK (SEQ ID NO: 48), or any of their fragments. In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, or 17 peptides. In some embodiments, the classifier comprises performance characterized by a receiver operating characteristic (ROC) curve having an average or median area under the curve (AUC) of at least 0.8. In some embodiments, the subject is suspected of having lung cancer. In some embodiments, evaluating comprises identifying that the biomarker indicates lung cancer. In some embodiments, the method further comprises the step of treating the subject for lung cancer or obtaining a lung nodule biopsy from the subject if the subject has lung cancer. In some embodiments, the method further comprises monitoring the subject if the subject does not have lung cancer. In some embodiments, the lung cancer comprises non-small cell lung cancer (NSCLC). In some embodiments, the biological fluid sample is obtained from a subject identified as having a lung nodule. In some embodiments, the method further comprises identifying the subject as having a lung nodule by performing medical imaging. In some embodiments, the classifier distinguishes between cancerous and non-cancerous lung nodules., 【0009】 In some aspects, this specification describes a method for detecting lung cancer, comprising: (a) obtaining a biomarker from a biological fluid sample of a subject; and (b) applying a classifier to the biomarker to evaluate for lung cancer, wherein the classifier has a performance characterized by a receiver operating characteristic (ROC) curve having an average or median area under the curve (AUC) of at least 0.7, and discriminates between biological fluid samples from subjects with and without lung cancer, and the biomarker is one of the following mRNA transcripts: ENSG00000155744.10, ENSG00000081052.14, ENSG00000173726.11, ENSG00000143995.20, ENSG00000108528.14, ENSG00000177427.13, ENSG00000163961.4, ENSG00000049130.16, ENSG00000008405.12, ENSG00000135090.14, ENSG00000151778.11, ENSG00000172116.23, ENSG00000144218.21, ENSG00000131196.18, ENSG00000129351.18, ENSG00000105518.14, ENSG00000182162.11, ENSG00000126368.6, ENSG00000176358.16, ENSG00000112599.9, ENSG00000142864.15, ENSG00000163159.15, ENSG00000165661.17, ENSG00000165661.18, ENSG00000007923.17, ENSG00000054116.12, ENSG00000113811.12, ENSG00000100644.17, ENSG00000133997.12, ENSG00000120925.16, ENSG00000110048.12, ENSG00000197863.9, ENSG00000174307.7, or ENSG00000109381.21, or one of the following peptides: LC(UniMod:4)PSGMYTEYIHSR (SEQ ID NO: 49), LCPSGMYTEYIHSR (SEQ ID NO: 139), NADLQVLKPEPELVYEDLR (SEQ ID NO: 50), ASTPGAAAQIQEVK (SEQ ID NO: 51),PYC(UniMod:4)NHPC(UniMod:4)YAAMFGPK(SEQ ID NO: 52), PYCNHPCYAAMFGPK(SEQ ID NO: 140), QLLQENEVQFLDK(SEQ ID NO: 53), AISAFHGSLSSSQPAEIITQSK(SEQ ID NO: 54), FEGIAC(UniMod:4)EISK(SEQ ID NO: 55), FEGIACEISK(SEQ ID NO: 141), FIINDWVK(SEQ ID NO: 56), YVGGQEHFAHLLILRDTK(SEQ ID NO: 57), SVGFHLPSR(SEQ ID NO: 58), GSPMEISLPIALSK(SEQ ID NO: 59), M(UniMod:35)VVSMTLGLHPWIANIDDTQYLAAK(SEQ ID NO: 60), MVVSMTLGLHPWIANIDDTQYLAAK(SEQ ID NO: 142), TVTAM(UniMod:35)DVVYALK(SEQ ID NO: 61), TVTAMDVVYALK(SEQ ID NO: 143), C(UniMod:4)SC(UniMod:4)DPGYELAPDKR(SEQ ID NO: 62), CSCDPGYELAPDKR(SEQ ID NO: 144), GNPTVEVDLHTAK(SEQ ID NO: 63), HLQLAIRNDEELNK(SEQ ID NO: 64), FQDGDLTLYQSNTILR(SEQ ID NO: 65), IRPNDFIPNVI(SEQ ID NO: 66), TKLEEHLEGIVNIFHQYSVRK(SEQ ID NO: 67), GDPEC(UniMod:4)HLFYNEQQEAR(SEQ ID NO: 68), GDPECHLFYNEQQEAR(SEQ ID NO: 145), ALNSIIDVYHK(SEQ ID NO: 69), DDPDAPLQPVTPLQLFEGR(SEQ ID NO: 70), KSEEENLFEIITADEVHYFLQAATPK(SEQ ID NO: 71), FPNGVQLSPAEDFVLVAETTMAR(SEQ ID NO: 72), LYFMHFNLESSYLC(UniMod:4)EYDYVK(SEQ ID NO: 73), LYFMHFNLESSYLCEYDYVK(SEQ ID NO: 146), LFDYC(UniMod:4)DIPLC(UniMod:4)ASSSFDC(UniMod:4)GK(SEQ ID NO: 74), LFDYCDIPLCASSSFDCGK(SEQ ID NO: 147), AEQC(UniMod:4)C(UniMod:4)EETASSISLHGK(SEQ ID NO: 75), AEQCCEETASSISLHGK(SEQ ID NO: 148)VALEGLRPTIPPGISPHVC (UniMod:4)K (SEQ ID NO:76), VALEGLRPTIPPGISPHVCK (SEQ ID NO:149), VWEQIDQMK (SEQ ID NO:77), FTDEEVDELYREAPIDK (SEQ ID NO:78), DTHFPIC (UniMod:4)IFC (UniMod:4)C (UniMod:4)GC (UniMod:4)C (UniMod:4)HR (SEQ ID NO:79), DTHFPICIFCCGCCHR (SEQ ID NO:150), RQDNEILIFWSK (SEQ ID NO:80), QDNEILIFWSK (SEQ ID NO:81), EVGTVLSQVYSK (SEQ ID NO:82), MVTALGTHWHPEHFC (UniMod:4)C (UniMod:4)VSC (UniMod:4)GEPFGDEGFHER (SEQ ID NO:83), MVTALGTHWHPEHFCCVSCGEPFGDEGFHER (SEQ ID NO:151), EVTFHC (UniMod:4)HEGYILHGAPK (SEQ ID NO:84), EVTFHCHEGYILHGAPK (SEQ ID NO:152), GAGGQSMSEAPTGDHAPAPTR (SEQ ID NO:85), DGSFSVVITGLR (SEQ ID NO:86), GISLNPEQWSQLK (SEQ ID NO:87), LVHVEEPHTETVR (SEQ ID NO:88), RVEPYGENFNK (SEQ ID NO:89), LDDC (UniMod:4)GLTEAR (SEQ ID NO:90), LDDCGLTEAR (SEQ ID NO:153), LVQAAQMLQSDPYSVPAR (SEQ ID NO:91), DFLGFYVVDSHR (SEQ ID NO:92), YGTC (UniMod:4)IYQGR (SEQ ID NO:93), YGTCIYQGR (SEQ ID NO:154), WLQEGGQEC (UniMod:4)EC (UniMod:4)K (SEQ ID NO:94), WLQEGGQECECK (SEQ ID NO:155), ASGPPVSELITK (SEQ ID NO:95), ELSDFISYLQR (SEQ ID NO:96), EGHVLQGPSVLK (SEQ ID NO:97), MNLASEPQEVLHIGSAHNR (SEQ ID NO:98), FLILPDMLK (SEQ ID NO:99), GISQEQMNEFR (SEQ ID NO:100), DPNHFRPAGLPEK (SEQ ID NO:101), VPSHLQAETLVGK (SEQ ID NO:102), NLHFLTTQEDYTLK (SEQ ID NO:103), SEAYNTFSER (SEQ ID NO:104)AVLDVFEEGTEASAATAVK (SEQ ID NO: 105), VIQYLAYVASSHK (SEQ ID NO: 106), ASYAQQPAESR (SEQ ID NO: 107), YLEESNFVHR (SEQ ID NO: 108), GSFTYFAPSNEAWDNLDSDIR (SEQ ID NO: 109), ALTDMPQM (UniMod:35)R (SEQ ID NO: 110), LAVNM (UniMod:35)VPFPR (SEQ ID NO: 111), TSC (UniMod:4)LLFMGR (SEQ ID NO: 112), QQQHLFGSNVTDC (UniMod:4)SGNFC (UniMod:4)LFR (SEQ ID NO: 113), ALTDMPQMR (SEQ ID NO: 156), LAVNMVPFPR (SEQ ID NO: 157), TSCLLFMGR (SEQ ID NO: 158), QQQHLFGSNVTDCSGNFCLFR (SEQ ID NO: 159), DYVSQFEGSALGK (SEQ ID NO: 114), DSITTWEILAVSMSDK (SEQ ID NO: 115), FC (UniMod:4)NIMGSSNGVDQEHFSNVVK (SEQ ID NO: 116), FCNIMGSSNGVDQEHFSNVVK (SEQ ID NO: 160), SEHPGLSIGDTAK (SEQ ID NO: 117), QFVEQHTPQLLTLVPR (SEQ ID NO: 118), NQDLAPNSAEQASILSLVTK (SEQ ID NO: 119), TDGALLVNAMFFK (SEQ ID NO: 120), DDFEGQLESDRFLLMSGGK (SEQ ID NO: 121), SIQC (UniMod:4)LTVHK (SEQ ID NO: 122), SIQCLTVHK (SEQ ID NO: 161), EDITQSAQHALR (SEQ ID NO: 123), VVAC (UniMod:4)TSAFLLWDPTK (SEQ ID NO: 124), VVACTSAFLLWDPTK (SEQ ID NO: 162), NYPMHVFAYR (SEQ ID NO: 125), MEEVEAMLLPETLK (SEQ ID NO: 126), ADVQAHGEGQEFSITC (UniMod:4)LVDEEEM (UniMod:35)K (SEQ ID NO: 127), ADVQAHGEGQEFSITCLVDEEEMK (SEQ ID NO: 163), DFALLSLQVPLK (SEQ ID NO: 128), LLIYAVLPTGDVIGDSAK (SEQ ID NO: 129), VDIVAINDPFIDLNYMVYMFQYDSTHGK (SEQ ID NO: 130), AEQINQAAGEASAVLAK (SEQ ID NO: 131)Any of TPAYYPNAGLIK (Accession No. 132), QGENGQMM (UniMod:35)SC (UniMod:4)TC (UniMod:4)LGNGK (Accession No. 133), QGENGQMMSCTCLGNGK (Accession No. 164), YWEMQPATFR (Accession No. 134), HGEYWLGNK (Accession No. 135), FVPAEMGTHTVSVK (Accession No. 136), NALGPGLSPELGPLPALR (Accession No. 137), or TKLEEHLEGIVNIFHQYSVR (Accession No. 138), and the following lipids: 1-palmitoyl-GPE (16:0), phosphatidylcholine (18:0 / 20:2, 20:0 / 18:2), linoleamide (18:2n6), linolenamide (18:3), 2-aminooctanoate, 1-linoleoyl-2-arachidonoyl-GPC (18:2 / 20:4n6), 1-palmitoylglycerol (16:0), 1-oleoyl-GPC (18:1), 1-linolenoyl-GPC (18:3), pregnanolone / allopregnanolone sulfate, sphingomyelin (d18:2 / 24:1, d18:1 / 24:2), myristoleamide (14:1), 1-linoleoylglycerol (18:2), 11beta-hydroxyandrostergone glucuronide, 2S,3R-dihydroxybutyrate, glycosyl-N-behenoyl-sphingosine (d18:1 / 22:0), 1-palmitoyl-2-linoleoyl-GPC (16:0 / 18:2), 1-stearoyl-2-arachidonoyl-GPS (18:0 / 20:4), 1-lignoceroyl-GPC (24:0), 3beta-hydroxy-5-cholestenate, 5alpha-androstan-3alpha,17beta-diol monosulfate (2), hexadecenedioate (C16:1-DC), myristamide (14:0), 1-stearoyl-GPE (18:0), 1-myristoyl-2-arachidonoyl-GPC (14:0 / 20:4), 1-arachidonyl-GPC (20:0), 4-hydroxy-2-oxoglutaric acid, nisinic acid (24:6n3), sphingomyelin (d17:1 / 16:0, d18:1 / 15:0, d16:1 / 17:0), 3-hydroxyoctanoate, 1-arachidonylglycerol (20:4),1-Stearoyl-2-oleoyl-GPS(18:0 / 18:1), 1-Eicosenoyl-GPE(20:1), Sphingosine, Glycoursodeoxycholic acid sulfate(1), 1-Stearoyl-2-linoleoyl-GPC(18:0 / 18:2), Erucate(22:1n9), Phosphoethanolamine, Ethiocholanolone glucuronide, Behenoyldihydrosphingomyelin(d18:0 / 22:0), Androste, Androsterol (3α,17α) monosulfate (2), isoursocholate, N-stearoyl-sphingosine (d18:1 / 18:0), margaramide (17:0), 1-eicosenoyl-GPC (20:1), tetrahydrocortisone glucuronide (5), linoleoyl carnitine (C18:2), hydroxypalmitoyl sphingomyelin (d18:1 / 16:0(OH)), or 1-eicosapentaenoyl-GPC (20:5), or one of the following metabolites: N-acetyl carnosine, indole lactate, lanthionine, 3-(4-hydroxyphenyl) lactate, hydantoin-5-propionate, urea, homoarginine, beta-cyano glutamate, S-1-pyrroline-5-carboxylate, aspartate, isovaleryl carnitine (C5), creatine, N-acetylglucosamine / N-acetylgalactosamine, galactonate, N-acetylneuraminic acid, 3-phosphoglycerate, bilirubin (E,Z or Z,E), retinol (vitamin A), heme, nicotinamide, carotenoid diol (1), bilirubin (Z,Z), 1-methylnicotinamide, alpha-ketoglutarate, xanthine, phenylacetyl carnitine, HWESASXX, 5-acetylamino-6-formylamino-3-methyluracil, 2-keto-3-deoxy-gluconate, iminodiacetate (IDA), 4-acetaminophen sulfate, caffeic acid sulfate, 2-hydroxyacetaminophen sulfate, 3-formylindole, X-18779, X-24473, X-23593, X-24307, X-24027, X-14939, X-12456, X-25790, X-17146, X-15220, X-12740, X-17765, X-25420, X-23639, X-12462, X-15728, or X-25422, a method is disclosed. In some embodiments, the biomarker is the following mRNA transcripts: ENSG00000155744.10, ENSG00000081052.14, ENSG00000173726.11, ENSG00000143995.20, ENSG00000108528.14, ENSG00000177427.13,Comprising any of ENSG00000163961.4, ENSG00000049130.16, ENSG00000008405.12, ENSG00000135090.14, ENSG00000151778.11, ENSG00000172116.23, ENSG00000144218.21, ENSG00000131196.18, ENSG00000129351.18, ENSG00000105518.14, ENSG00000182162.11, ENSG00000126368.6, ENSG00000176358.16, ENSG00000112599.9, ENSG00000142864.15, ENSG00000163159.15, ENSG00000165661.17, ENSG00000165661.18, ENSG00000007923.17, ENSG00000054116.12, ENSG00000113811.12, ENSG00000100644.17, ENSG00000133997.12, ENSG00000120925.16, ENSG00000110048.12, ENSG00000197863.9, ENSG00000174307.7, or ENSG00000109381.21. In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, or 28 mRNA transcripts. In some embodiments, the biomarker is the following peptide LC (UniMod:4) PSGMYTEYIHSR (SEQ ID NO: 49), LCPSGMYTEYIHSR (SEQ ID NO: 139), NADLQVLKPEPELVYEDLR (SEQ ID NO: 50), ASTPGAAAQIQEVK (SEQ ID NO: 51), PYC (UniMod:4) NHPC (UniMod:4) YAAMFGPK (SEQ ID NO: 52), PYCNHPCYAAMFGPK (SEQ ID NO: 140), QLLQENEVQFLDK (SEQ ID NO: 53), AISAFHGSLSSSQPAEIITQSK (SEQ ID NO: 54), FEGIAC (UniMod:4) EISK (SEQ ID NO: 55), FEGIACEISK (SEQ ID NO: 141), FIINDWVK (SEQ ID NO: 56),YVGGQEHFAHLLILRDTK (SEQ ID NO: 57), SVGFHLPSR (SEQ ID NO: 58), GSPMEISLPIALSK (SEQ ID NO: 59), M (UniMod:35)VVSMTLGLHPWIANIDDTQYLAAK (SEQ ID NO: 60), MVVSMTLGLHPWIANIDDTQYLAAK (SEQ ID NO: 142), TVTAM (UniMod:35)DVVYALK (SEQ ID NO: 61), TVTAMDVVYALK (SEQ ID NO: 143), C (UniMod:4)SC (UniMod:4)DPGYELAPDKR (SEQ ID NO: 62), CSCDPGYELAPDKR (SEQ ID NO: 144), GNPTVEVDLHTAK (SEQ ID NO: 63), HLQLAIRNDEELNK (SEQ ID NO: 64), FQDGDLTLYQSNTILR (SEQ ID NO: 65), IRPNDFIPNVI (SEQ ID NO: 66), TKLEEHLEGIVNIFHQYSVRK (SEQ ID NO: 67), GDPEC (UniMod:4)HLFYNEQQEAR (SEQ ID NO: 68), GDPECHLFYNEQQEAR (SEQ ID NO: 145), ALNSIIDVYHK (SEQ ID NO: 69), DDPDAPLQPVTPLQLFEGR (SEQ ID NO: 70), KSEEENLFEIITADEVHYFLQAATPK (SEQ ID NO: 71), FPNGVQLSPAEDFVLVAETTMAR (SEQ ID NO: 72), LYFMHFNLESSYLC (UniMod:4)EYDYVK (SEQ ID NO: 73), LYFMHFNLESSYLCEYDYVK (SEQ ID NO: 146), LFDYC (UniMod:4)DIPLC (UniMod:4)ASSSFDC (UniMod:4)GK (SEQ ID NO: 74), LFDYCDIPLCASSSFDCGK (SEQ ID NO: 147), AEQC (UniMod:4)C (UniMod:4)EETASSISLHGK (SEQ ID NO: 75), AEQCCEETASSISLHGK (SEQ ID NO: 148), VALEGLRPTIPPGISPHVC (UniMod:4)K (SEQ ID NO: 76), VALEGLRPTIPPGISPHVCK (SEQ ID NO: 149), VWEQIDQMK (SEQ ID NO: 77), FTDEEVDELYREAPIDK (SEQ ID NO: 78), DTHFPIC (UniMod:4)IFC (UniMod:4)C (UniMod:4)GC (UniMod:4)C (UniMod:4)HR (SEQ ID NO: 79),DTHFPICIFCCGCCHR (SEQ ID NO: 150), RQDNEILIFWSK (SEQ ID NO: 80), QDNEILIFWSK (SEQ ID NO: 81), EVGTVLSQVYSK (SEQ ID NO: 82), MVTALGTHWHPEHFC (UniMod:4)C (UniMod:4)VSC (UniMod:4)GEPFGDEGFHER (SEQ ID NO: 83), MVTALGTHWHPEHFCCVSCGEPFGDEGFHER (SEQ ID NO: 151), EVTFHC (UniMod:4)HEGYILHGAPK (SEQ ID NO: 84), EVTFHCHEGYILHGAPK (SEQ ID NO: 152), GAGGQSMSEAPTGDHAPAPTR (SEQ ID NO: 85), DGSFSVVITGLR (SEQ ID NO: 86), GISLNPEQWSQLK (SEQ ID NO: 87), LVHVEEPHTETVR (SEQ ID NO: 88), RVEPYGENFNK (SEQ ID NO: 89), LDDC (UniMod:4)GLTEAR (SEQ ID NO: 90), LDDCGLTEAR (SEQ ID NO: 153), LVQAAQMLQSDPYSVPAR (SEQ ID NO: 91), DFLGFYVVDSHR (SEQ ID NO: 92), YGTC (UniMod:4)IYQGR (SEQ ID NO: 93), YGTCIYQGR (SEQ ID NO: 154), WLQEGGQEC (UniMod:4)EC (UniMod:4)K (SEQ ID NO: 94), WLQEGGQECECK (SEQ ID NO: 155), ASGPPVSELITK (SEQ ID NO: 95), ELSDFISYLQR (SEQ ID NO: 96), EGHVLQGPSVLK (SEQ ID NO: 97), MNLASEPQEVLHIGSAHNR (SEQ ID NO: 98), FLILPDMLK (SEQ ID NO: 99), GISQEQMNEFR (SEQ ID NO: 100), DPNHFRPAGLPEK (SEQ ID NO: 101), VPSHLQAETLVGK (SEQ ID NO: 102), NLHFLTTQEDYTLK (SEQ ID NO: 103), SEAYNTFSER (SEQ ID NO: 104), AVLDVFEEGTEASAATAVK (SEQ ID NO: 105), VIQYLAYVASSHK (SEQ ID NO: 106), ASYAQQPAESR (SEQ ID NO: 107), YLEESNFVHR (SEQ ID NO: 108), GSFTYFAPSNEAWDNLDSDIR (SEQ ID NO: 109), ALTDMPQM (UniMod:35)R (SEQ ID NO: 110), LAVNM (UniMod:35)VPFPR (SEQ ID NO: 111),TSC (UniMod:4) LLFMGR (SEQ ID NO: 112), QQQHLFGSNVTDC (UniMod:4) SGNFC (UniMod:4) LFR (SEQ ID NO: 113), ALTDMPQMR (SEQ ID NO: 156), LAVNMVPFPR (SEQ ID NO: 157), TSCLLFMGR (SEQ ID NO: 158), QQQHLFGSNVTDCSGNFCLFR (SEQ ID NO: 159), DYVSQFEGSALGK (SEQ ID NO: 114), DSITTWEILAVSMSDK (SEQ ID NO: 115), FC (UniMod:4) NIMGSSNGVDQEHFSNVVK (SEQ ID NO: 116), FCNIMGSSNGVDQEHFSNVVK (SEQ ID NO: 160), SEHPGLSIGDTAK (SEQ ID NO: 117), QFVEQHTPQLLTLVPR (SEQ ID NO: 118), NQDLAPNSAEQASILSLVTK (SEQ ID NO: 119), TDGALLVNAMFFK (SEQ ID NO: 120), DDFEGQLESDRFLLMSGGK (SEQ ID NO: 121), SIQC (UniMod:4) LTVHK (SEQ ID NO: 122), SIQCLTVHK (SEQ ID NO: 161), EDITQSAQHALR (SEQ ID NO: 123), VVAC (UniMod:4) TSAFLLWDPTK (SEQ ID NO: 124), VVACTSAFLLWDPTK (SEQ ID NO: 162), NYPMHVFAYR (SEQ ID NO: 125), MEEVEAMLLPETLK (SEQ ID NO: 126), ADVQAHGEGQEFSITC (UniMod:4) LVDEEEM (UniMod:35) K (SEQ ID NO: 127), ADVQAHGEGQEFSITCLVDEEEMK (SEQ ID NO: 163), DFALLSLQVPLK (SEQ ID NO: 128), LLIYAVLPTGDVIGDSAK (SEQ ID NO: 129), VDIVAINDPFIDLNYMVYMFQYDSTHGK (SEQ ID NO: 130), AEQINQAAGEASAVLAK (SEQ ID NO: 131), TPAYYPNAGLIK (SEQ ID NO: 132), QGENGQMM (UniMod:35) SC (UniMod:4) TC (UniMod:4) LGNGK (SEQ ID NO: 133), QGENGQMMSCTCLGNGK (SEQ ID NO: 164), YWEMQPATFR (SEQ ID NO: 134), HGEYWLGNK (SEQ ID NO: 135), FVPAEMGTHTVSVK (SEQ ID NO: 136), NALGPGLSPELGPLPALR (SEQ ID NO: 137)or contains any of TKLEEHLEGIVNIFHQYSVR (SEQ ID NO: 138). In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, or 28 peptides. In some embodiments, the biomarker is the following lipid, 1-palmito, Ile-GPE(16:0), Phosphatidylcholine(18:0 / 20:2, 20:0 / 18:2), Linoleamide(18:2n6), Linolenamide(18:3), 2-Amino octanoate, 1-Linoleoyl-2-arachidonoyl-GPC(18:2 / 20:4n6), 1-Palmitoyl glycerol(16:0), 1-Oleoyl-GPC(18:1), 1-Linolenoyl-GPC(18:3), Pregnanolone / Allopregnanolone sulfate, Sphingomyelin(d18:2 / 24:1, d18:1 / 24:2), Myristoleamide(14:1), 1-Linoleoyl glycerol(18:2), 11 beta-Hydroxy androsterone glucuronide, 2S,3R-Dihydroxy butyrate, Glycosyl-N-behenoyl-sphingosine(d18:1 / 22:0), 1-Palmitoyl-2-linoleoyl-GPC(16:0 / 18:2), 1-Stearoyl-2-arachidonoyl-GPS(18:0 / 20:4), 1-Lignoceroyl-GPC(24:0), 3 beta-Hydroxy-5-cholestenate, 5 alpha-Androstan-3 alpha,17 beta-diol monosulfate(2), Hexadecenedioate(C16:1-DC), Myristamide(14:0), 1-Stearoyl-GPE(18:0), 1-Myristoyl-2-arachidonoyl-GPC(14:0 / 20:4), 1-Arachidonyl-GPC(20:0), 4-Hydroxy-2-oxoglutaric acid, Nisinic acid(24:6n3), Sphingomyelin(d17:1 / 16:0, d18:1 / 15:0, d16:1 / 17:0), 3-Hydroxy octanoate, 1-Arachidonyl glycerol(20:4), 1-Stearoyl-2-oleoyl-GPS(18:0 / 18:1), 1-Eicosenoyl-GPE(20:1), Sphingosine, Glycoursodeoxycholic acid sulfate(1), 1-Stearoyl-2-linoleoyl-GPC(18:0 / 18:2), Erucate(22:1n9), Phosphoethanolamine, Ethiocholanolone glucuronide, Behenoyl dihydrosphingomyelin(d18:0 / 22:0), Androstenediol(3 alpha,It contains any one of 17α)-monosulfate(2), isoursodeoxycholate, N-stearoyl-sphingosine (d18:1 / 18:0), margaramide(17:0), 1-eicosenoyl-GPC(20:1), tetrahydrocortisone glucuronide(5), linoleoyl carnitine (C18:2), hydroxypalmitoyl sphingomyelin (d18:1 / 16:0(OH)), or 1-eicosapentaenoyl-GPC(20:5). In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, or 28 lipids. In some embodiments, the biomarker is the following metabolite: N-acetyl carnosine, indole lactate, lanthionine, 3-(4-hydroxyphenyl)lactate, hydantoin-5-propionate, urea, homoarginine, beta-cyano glutamate, S-1-pyrroline-5-carboxylate, aspartate, isovaleryl carnitine (C5), creatine, N-acetylglucosamine / N-acetylgalactosamine, galactonate, N-acetylneuraminate, 3-phosphoglycerate, bilirubin (E,Z or Z,E), retinol (vitamin A), heme, nicotinamide, carotenoid diol(1), bilirubin (Z,Z), 1-methylnicotinamide, alpha-ketoglutarate, xanthine, phenylacetylcarnitine, HWESASXX, 5-acetylamino-6-formylamino-3-methyluracil, 2-keto-3-deoxy-gluconate, iminodiacetate (IDA), 4-acetaminophen sulfate, caffeic acid sulfate, 2-hydroxyacetaminophen sulfate, 3-formylindole, X-18779, X-24473, X-23593, X-24307, X-24027, X-14939, X-12456, X-25790, X-17146, X-15220, X-12740, X-17765, X-25420, X-23639, X-12462, X-15728, or X-25422. In some embodiments, the biomarker comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, or 28 metabolites. In some embodiments, the classifier comprises a performance characterized by a receiver operating characteristic (ROC) curve having an average or median area under the curve (AUC) of at least 0.8. In some embodiments, the subject is suspected of having lung cancer. In some embodiments, evaluating comprises identifying a biomarker indicative of lung cancer. In some embodiments, the method comprises administering a lung cancer treatment to the subject if the subject has lung cancer. In some embodiments, the method comprises monitoring the subject if the subject does not have lung cancer. In some embodiments, the lung cancer comprises non-small cell lung cancer. In some embodiments, the lung cancer comprises stage 1, 2, or 3 lung cancer. In some embodiments, the lung cancer comprises stage 4 lung cancer., 【0010】 In some aspects, the present specification discloses a multi-omics method. This method includes obtaining multi-omics data generated from one or more biological fluid samples collected from a subject suspected of having a disease state, where the multi-omics data includes proteomics measurements and nucleic acid sequencing measurements; applying a classifier to the multi-omics data to evaluate the disease state; and any one of (i)-(iv): (i) the proteomics measurements are generated after the samples of the one or more biological fluid samples have undergone a concentration protocol that concentrates proteins or peptides without concentrating another protein or peptide, (ii) the proteomics measurements are generated based on the amount of protein or peptide added to the samples of the one or more biological fluid samples, or (iii) the classifier includes performance characteristics including an area under the curve (AUC) of the receiver operating characteristic (ROC) curve with an average or median of at least 0.9, determined in a dataset derived from a randomized controlled trial of at least 20 subjects with the disease state and more than 20 control subjects without the disease state, or (iv) the evaluation includes selecting a cancer treatment based on the multi-omics data, and the proteomics measurements are generated using mass spectrometry. In some aspects, the proteomics measurements are generated after the samples of the one or more biological fluid samples have undergone a concentration protocol that concentrates some proteins without concentrating other proteins. In some aspects, the proteomics measurements are generated from proteins adsorbed to nanoparticles. In some aspects, the proteomics measurements are generated based on the amount of protein added to the samples of the one or more biological fluid samples. In some aspects, the protein added to the sample is labeled. In some aspects, the nucleic acid sequencing measurements include mRNA sequencing measurements. In some aspects, the nucleic acid sequencing measurements include mRNA sequencing measurements and miRNA sequencing measurements. In some aspects, the multi-omics data includes measurements of more than 45 peptide or protein groups.In some embodiments, the performance of the classifier is improved by at least 4% compared to when the classifier is applied to only one type of omics data, and the performance is determined in a dataset derived from a randomized controlled trial of more than 25 subjects with a disease state and more than 25 control subjects without a disease state, and includes sensitivity at a given specificity. In some embodiments, the classifier is characterized by an average area under the curve (AUC) of at least 0.9 of the receiver operating characteristic (ROC) curve, which is determined in a dataset derived from a randomized controlled trial of at least 20 subjects with a disease state and more than 20 control subjects without a disease state. In some embodiments, applying the classifier to multi-omics data to evaluate a disease state includes applying a first classifier to proteomics measurements to generate a first label corresponding to the presence, absence, or likelihood of a disease state, applying a second classifier to nucleic acid sequencing measurements to generate a second label corresponding to the presence, absence, or likelihood of a disease state, and (a), (b), or (c): (a) the unweighted average of the first and second labels, (b) the weighted average of the first and second labels, or (c) evaluating the disease state based on a majority vote score based on the first and second labels. Some embodiments include evaluating the disease state based on the weighted average of the first and second labels, and the weighted average is generated by assigning weights to the results of the first and second classifiers based on the area under the ROC curve, the area under the precision-recall curve, accuracy, precision, recall, sensitivity, F1-score, specificity, or a combination thereof. In some embodiments, applying the classifier to multi-omics data to evaluate a disease state includes obtaining a subset of features from among the proteomics measurements; obtaining at least one subset of features from among the nucleic acid sequencing measurements; pooling a subset of features from the first omics data and at least one subset of features from the second omics data into the obtained pooled features; and evaluating the disease state based on the pooled features.In some embodiments, obtaining a subset of features from the first or second omics data includes obtaining top features based on univariate data. In some embodiments, the classifier is trained using deep learning, hierarchical cluster analysis, principal component analysis, partial least squares discriminant analysis, random forest classification analysis, support vector machine analysis, k-nearest neighbor analysis, naive Bayes analysis, K-means clustering analysis, or hidden Markov analysis. In some embodiments, the multi-omics data further includes metabolomics data. In some embodiments, the disease state includes cancer. In some embodiments, the cancer is selected from the group consisting of lung cancer, pancreatic cancer, breast cancer, colorectal cancer, liver cancer, and ovarian cancer. In some embodiments, the evaluation includes selecting a cancer treatment based on the multi-omics data. Some embodiments include administering chemotherapy, a medicament, radiation, or a surgical cancer treatment to the subject based on the evaluation. In some embodiments, the one or more biological fluid samples include blood, serum, or plasma samples. In some embodiments, the subject is human.In some aspects, the present specification is to obtain multi-omics data generated from one or more blood, serum, or plasma samples taken from a human subject suspected of having cancer, the multi-omics data including proteomic measurements and RNA sequencing measurements; applying a classifier to the multi-omics data to evaluate cancer; selecting or administering a cancer treatment method to the subject based on the evaluation; and any one of (i)-(iii): (i) the proteomic measurements are generated after the samples of one or more blood, serum, or plasma samples are enriched by an affinity reagent for a protein or peptide, (ii) the proteomic measurements are generated based on the amount of labeled protein or peptide added to the samples of one or more blood, serum, or plasma samples, or (iii) the classifier includes performance characteristics including an average area under the curve (AUC) of a receiver operating characteristic (ROC) curve of at least 0.9 determined in a held-out data set derived from a randomized controlled trial of at least 25 subjects with a disease state and more than 25 control subjects without a disease state. A multi-omics method is disclosed. In some embodiments, the proteomic measurements are generated after the samples of one or more blood, serum, or plasma samples are enriched by an affinity reagent. In some embodiments, the proteomic measurements are generated based on the amount of labeled protein added to the samples of one or more blood, serum, or plasma samples. In some embodiments, the classifier is characterized by an average area under the curve (AUC) of a receiver operating characteristic (ROC) curve of at least 0.9 determined in a data set derived from a randomized controlled trial of at least 25 subjects with a disease state and more than 25 control subjects without a disease state. 【0011】 In some aspects, the present specification is to obtain multi-omics data generated from one or more biological fluid samples collected from a subject, where the multi-omics data includes first omics data including proteomics data, metabolomics data, transcriptomics data, or genomics data, and second omics data including proteomics data, metabolomics data, transcriptomics data, or genomics data different from the first omics data; and using a first classifier to assign a first label including the presence, absence, or likelihood of a disease state to the first omics data, using a second classifier to assign a second label including the presence, absence, or likelihood of a disease state to the second omics data based on the first and second labels, and identifying the multi-omics data as indicating or not indicating a disease state. In some aspects, the first omics data includes proteomics data, and the second omics data includes metabolomics data, transcriptomics data, or genomics data. In some aspects, the proteomics data is generated by contacting one of the biological fluid samples with particles such that the particles adsorb biomolecules including proteins. In some aspects, the particles include carboxylate particles, polyacrylic acid particles, dextran particles, polystyrene particles, dimethylamine particles, amino particles, silica particles, or N-(3-trimethoxysilylpropyl)diethylenetriamine particles. In some aspects, the particles include a group of nanoparticles that are physiochemically different. In some aspects, the proteomics data is generated using mass spectrometry, chromatography, liquid chromatography, high performance liquid chromatography, solid phase chromatography, lateral flow assay, immunoassay, enzyme-linked immunosorbent assay, western blot, dot blot, or immunostaining, or a combination thereof.In some embodiments, genomic data or transcriptomic data is generated by sequencing, microarray analysis, hybridization, polymerase chain reaction, electrophoresis, or a combination thereof. In some embodiments, the second omics data includes transcriptomic data. In some embodiments, the transcriptomic data includes mRNA or microRNA expression data. In some embodiments, the second omics data includes genomic data. In some embodiments, the genomic data includes DNA sequence data or epigenetic data. In some embodiments, identifying multi-omics data as indicative or not indicative of a disease state includes identifying the multi-omics data as indicative or not indicative of a disease state based on either a first label or a second label. In some embodiments, identifying multi-omics data as indicative or not indicative of a disease state includes generating or obtaining a majority vote score based on the first and second labels. In some embodiments, identifying multi-omics data as indicative or not indicative of a disease state includes generating or obtaining a weighted average of the first and second labels. Some embodiments include assigning weights to first and second classifiers based on the area under the receiver operating characteristic (ROC) curve, the area under the precision-recall curve, accuracy, precision, recall, sensitivity, F1-score, specificity, or a combination thereof, thereby obtaining a weighted average. In some embodiments, the first omics data is generated from a first biological fluid sample of biological fluid samples, and the second omics data is generated from a second biological fluid sample of biological fluid samples. In some embodiments, the first biological fluid sample is collected in a first container containing a first collection component including heparin, ethylenediaminetetraacetic acid (EDTA), citrate, or an anti-lysing agent, and the second biological fluid sample is collected in a second container containing a second collection component including heparin, EDTA, citrate, or an anti-lysing agent different from the first collection component. In some embodiments, the multi-omics data further includes third omics data including a third omics data type.The third omics data may include an omics data type or subtype different from the first and second omics data. Some embodiments include using a third classifier to assign a third label corresponding to the presence, absence, or likelihood of a disease state to the third omics data. In some embodiments, identifying the multi-omics data as indicative or not indicative of a disease state includes identifying the multi-omics data as indicative or not indicative of a disease state based on a combination of the first, second, and third labels. Some embodiments include using a third classifier to assign a third label including the presence, absence, or likelihood of a disease state to third omics data different from the first and second omics data, and identifying the multi-omics data as indicative or not indicative of a disease state based on the first and second labels includes identifying the multi-omics data as indicative or not indicative of a disease state based on the first, second, and third labels. In some embodiments, the first omics data type includes proteomics data, the second omics data type includes mRNA transcriptomics data, and the third omics data type includes microRNA transcriptomics data (i.e., microRNA data). Some embodiments include transmitting or outputting information related to the identification. Some embodiments include recommending treatment of the disease state. 【0012】 In some aspects, this specification discloses a method that includes obtaining combined data including two, three, or four of proteomic data, metabolomic data, transcriptomic data, or genomic data generated from one or more biological fluid samples from a subject; and using a classifier to identify the combined data as indicative of or not indicative of one or more disease states. In some aspects, the one or more biological fluid samples include 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10, or more biological fluid samples. In some aspects, the combined data is generated simultaneously. In some aspects, simultaneous data generation includes assaying two, three, or four of proteomic data, metabolomic data, transcriptomic data, or genomic data simultaneously. In some aspects, simultaneous data generation includes assaying two, three, or four of proteomic data, metabolomic data, transcriptomic data, or genomic data at separate positions on an assay substrate. In some aspects, the separate positions include separate wells and the assay substrate includes an assay plate. In some aspects, the one or more biological fluid samples include two or more of whole blood samples, plasma samples, serum samples, or urine samples. In some aspects, the proteomic data is generated from one of the one or more biological fluid samples. In some aspects, the metabolomic data is generated from a biological fluid sample or from an additional biological fluid sample of the one or more biological fluid samples, and the proteomic data and the metabolomic data are combined to obtain the combined data. In some aspects, the classifier identifies the combined data as indicative of or not indicative of one or more disease states with higher sensitivity or specificity than proteomic data, metabolomic data, transcriptomic data, or genomic data alone. In some aspects, the classifier includes features selected from proteomic data, metabolomic data, genomic data, or transcriptomic data.In some embodiments, the classifier includes features selected from a combination of proteomics data, metabolomics data, genomics data, or transcriptomics data. In some embodiments, the classifier includes a plurality of classifiers. In some embodiments, the plurality of classifiers includes 2, 3, or 4, or more classifiers. In some embodiments, the plurality of classifiers separately includes features selected from proteomics data, metabolomics data, genomics data, transcriptomics data, or combinations thereof. In some embodiments, using a classifier to identify combined data as indicative or not indicative of one or more disease states includes using a plurality of classifiers to identify combined data as indicative or not indicative of one or more disease states. In some embodiments, using a classifier to identify combined data as indicative or not indicative of one or more disease states includes selecting the output of any one of the plurality of classifiers. In some embodiments, using a classifier to identify combined data as indicative or not indicative of one or more disease states includes a majority vote across a plurality of classifiers. In some embodiments, using a classifier to identify combined data as indicative or not indicative of one or more disease states includes a majority vote across a subset of the plurality of classifiers. In some embodiments, using a classifier to identify combined data as indicative or not indicative of one or more disease states includes a weighted average of a plurality of classifiers. In some embodiments, using a classifier to identify combined data as indicative or not indicative of one or more disease states includes a weighted average of a subset of the plurality of classifiers. In some embodiments, the weights of the weighted average are assigned based on the area under the receiver operating characteristic (ROC) curve. In some embodiments, the weights of the weighted average are assigned based on the area under the precision-recall curve. In some embodiments, the weights of the weighted average are assigned based on accuracy. In some embodiments, the weights of the weighted average are assigned based on precision. In some embodiments, the weights of the weighted average are assigned based on recall. In some embodiments, the weights of the weighted average are assigned based on sensitivity.In some aspects, the weights of the weighted average are assigned based on the F1-score. In some aspects, the weights of the weighted average are assigned based on the specificity. 【0013】 In some aspects, this specification discloses a method that includes obtaining proteomics data generated from a biological fluid sample of a subject; obtaining metabolomics data, transcriptomics data, or genomics data generated from a biological fluid sample of the subject or from an additional biological fluid sample, where the proteomics data is combined with the metabolomics data, transcriptomics data, or genomics data to obtain combined data; and using a classifier to identify the combined data as indicative or non-indicative of one or more disease states. In some aspects, the proteomics data is generated by contacting a biological fluid sample of the subject with particles such that the particles adsorb biomolecules that include proteins. Some aspects include contacting a biological fluid sample of the subject with particles such that the particles adsorb biomolecules. Some aspects include analyzing the biomolecules adsorbed to the particles to generate proteomics data. Some aspects include analyzing the biological fluid sample or an additional biological fluid sample to generate metabolomics data. Some aspects include using a classifier to identify the combined data as indicative or non-indicative of one or more disease states. In some aspects, the proteomics data is generated by measuring a reading indicative of the presence, absence, or amount of a biomolecule. In some aspects, the proteomics data is generated using mass spectrometry, chromatography, liquid chromatography, high performance liquid chromatography, solid phase chromatography, lateral flow assay, immunoassay, enzyme-linked immunosorbent assay, western blot, dot blot, or immunostaining, or a combination thereof. In some aspects, the proteomics data is generated using mass spectrometry. In some aspects, the proteins include secreted proteins. In some aspects, the particles include nanoparticles. In some aspects, the particles include lipid particles, metal particles, silica particles, or polymer particles.In some embodiments, the particles include carboxylate particles, polyacrylic acid particles, dextran particles, polystyrene particles, dimethylamine particles, amino particles, silica particles, or N-(3-trimethoxysilylpropyl)diethylenetriamine particles. In some embodiments, the particles include physiochemically different groups of nanoparticles. In some embodiments, the metabolomics data is generated from a biofluid sample different from the proteomics data. In some embodiments, the metabolomics data is generated using mass spectrometry, electrophoresis, colorimetric assay, fluorescence assay, chromatography, liquid chromatography, high performance liquid chromatography, solid phase chromatography, lateral flow assay, immunoassay, or combinations thereof. In some embodiments, the metabolomics data is generated using mass spectrometry. In some embodiments, the metabolomics data is generated from the same biofluid sample as the proteomics data. In some embodiments, the metabolomics data is generated by analyzing an analyte adsorbed to the particles. In some embodiments, the metabolomics data includes lipid metabolite data, carbohydrate metabolite data, vitamin metabolite data, cofactor metabolite data, or combinations thereof. In some embodiments, the biofluid sample includes a blood sample, plasma sample, or serum sample. In some embodiments, an additional biofluid sample is collected from the subject in a container separate from the biofluid sample. In some embodiments, the combined data is generated from 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more samples. In some embodiments, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more samples are collected separately in 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more containers. In some embodiments, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more containers include a plurality of components in addition to the sample. In some embodiments, the biofluid sample and the additional biofluid sample are collected in separate containers, and these separate containers contain different components. In some embodiments, the first of the separate containers includes a first component different from a second component in the second of the separate containers.In some embodiments, the biological fluid sample comprises serum; is collected in a container containing ethylenediaminetetraacetic acid (EDTA), citrate, or heparin; or contains a preservative that prevents cell lysis. In some embodiments, the biological fluid sample is collected in a container containing ethylenediaminetetraacetic acid (EDTA). In some embodiments, the additional biological fluid sample comprises a blood sample, a plasma sample, or a serum sample. In some embodiments, the additional biological fluid sample is processed to obtain cell-free DNA or to obtain RNA. Some embodiments include obtaining genomic data or transcriptomic data generated from a biological fluid sample from a subject, from an additional biological fluid sample, or from a third biological fluid sample. In some embodiments, the combined data further comprises genomic data or transcriptomic data. Some embodiments include analyzing a biological fluid sample, an additional biological fluid sample, or a third biological fluid sample to generate genomic data or transcriptomic data. In some embodiments, the third biological fluid sample comprises a blood sample, a plasma sample, or a serum sample. In some embodiments, the third biological fluid sample is processed to obtain cell-free DNA or to obtain RNA. Some embodiments include using a classifier to identify combined data as indicative of or not indicative of one or more disease states. In some embodiments, the genomic data or transcriptomic data is generated by measuring read values indicative of the presence, absence, or amount of nucleic acid. In some embodiments, the genomic data or transcriptomic data is generated by sequencing, microarray analysis, hybridization, polymerase chain reaction, electrophoresis, or a combination thereof. In some embodiments, the genomic data or transcriptomic data is generated from a biological fluid sample different from metabolomic data. In some embodiments, the genomic data or transcriptomic data is generated from the same biological fluid sample as metabolomic data.In some embodiments, the genomics data or transcriptomics data is generated from a biological fluid sample different from the p data. In some embodiments, the genomics data or transcriptomics data is generated from the same biological fluid sample as the proteomics data. In some embodiments, the genomics data or transcriptomics data is generated by analyzing nucleic acids adsorbed to particles. In some embodiments, the genomics data or transcriptomics data includes genomics data. In some embodiments, the genomics data includes DNA sequence data. In some embodiments, the genomics data includes DNA polymorphism data. In some embodiments, the genomics data includes epigenetic data. In some embodiments, the genomics data includes DNA methylation data. In some embodiments, the epigenetic data includes histone modification data. In some embodiments, the histone modification data includes acetylation data, methylation data, ubiquitination data, phosphorylation data, SUMOylation data, ribosylation data, or citrullination data. In some embodiments, the genomics data or transcriptomics data includes transcriptomics data. In some embodiments, the transcriptomics data includes RNA sequence data. In some embodiments, the transcriptomics data includes RNA expression data. In some embodiments, the transcriptomics data includes mRNA, tRNA, rRNA, microRNA, snRNA, snoRNA, or lncRNA expression data. In some embodiments, the transcriptomics data includes mRNA expression data. In some embodiments, the transcriptomics data includes microRNA expression data. In some embodiments, the classifier includes features for identifying combined data as indicative of one or more disease states. In some embodiments, the features include control protein measurements, control metabolite measurements, control nucleic acid measurements, mass spectra, m / z ratios, chromatography results, immunoassay results, light or fluorescence intensity, or sequence information.In some aspects, the classifier is trained using deep learning, hierarchical cluster analysis, principal component analysis, partial least squares discriminant analysis, random forest classification analysis, support vector machine analysis, k-nearest neighbor analysis, naive Bayes analysis, K-means clustering analysis, or hidden Markov analysis. In some aspects, one or more disease states include one or more cancers. In some aspects, one or more cancers include lung cancer, breast cancer, prostate cancer, colorectal cancer, colon cancer, melanoma, bladder cancer, lymphoma, leukemia, kidney cancer, uterine cancer, pancreatic cancer, or combinations thereof. In some aspects, the classifier differentiates one or more disease states. In some aspects, the classifier differentiates lung cancer, colon cancer, and pancreatic cancer. In some aspects, the classifier differentiates lung cancer, colon cancer, and pancreatic cancer. In some aspects, lung cancer includes non-small cell lung cancer (NSCLC). Some aspects include generating a report based on the use of a classifier to identify combined data as indicative or non-indicative of one or more disease states. In some aspects, the report includes a biological fluid or an indication or metric that the subject may include one or more disease states. Some aspects include outputting or transmitting the report. In some aspects, the report is used by a healthcare provider in making a diagnosis, providing medical advice, or offering treatment for at least one of the one or more disease states. Some aspects include identifying combined data as indicative of one or more disease states. In some aspects, one or more disease states include cancer, and further include recommending cancer treatment to the subject if the combined data is identified as indicative of cancer. In some aspects, one or more disease states include cancer, and further include subjecting the subject to cancer treatment if the combined data is identified as indicative of cancer. In some aspects, cancer treatment includes chemotherapy, radiation therapy, ablation therapy, embolization, or surgery. Some aspects include using the classifier to identify combined data as indicative of a first disease state among one or more disease states and non-indicative of a second disease state among one or more disease states.Some aspects include treating or recommending treatment for a first disease state rather than a second disease state. Some aspects include identifying combined data as not indicative of one or more disease states. Some aspects include observing a subject without providing treatment to the subject when the combined data is identified as not indicative of one or more disease states. In some aspects, observing the subject without providing treatment includes analyzing biomolecules in a biological fluid sample later obtained from the subject. In some aspects, the subject is a mammal. In some aspects, the subject is a human. In some aspects, the classifier includes features selected from proteomics data, metabolomics data, genomics data, or transcriptomics data. Some. In some embodiments, the classifier includes features selected from a combination of proteomics data, metabolomics data, genomics data, or transcriptomics data. In some embodiments, the classifier includes a plurality of classifiers. In some embodiments, the plurality of classifiers includes two, three, or four, or more classifiers. In some embodiments, the plurality of classifiers separately includes features selected from proteomics data, metabolomics data, genomics data, transcriptomics data, or combinations thereof. In some embodiments, identifying the combined data as indicative of or not indicative of one or more disease states using a classifier includes identifying the combined data as indicative of or not indicative of one or more disease states using a plurality of classifiers. In some embodiments, identifying the combined data as indicative of or not indicative of one or more disease states using a classifier includes selecting the output of any one of the plurality of classifiers. In some embodiments, identifying the combined data as indicative of or not indicative of one or more disease states using a classifier includes a majority vote across the plurality of classifiers. In some embodiments, identifying the combined data as indicative of or not indicative of one or more disease states using a classifier includes a majority vote across a subset of the plurality of classifiers. In some embodiments, identifying the combined data as indicative of or not indicative of one or more disease states using a classifier includes a weighted average of the plurality of classifiers. In some embodiments, identifying the combined data as indicative of or not indicative of one or more disease states using a classifier includes a weighted average of a subset of the plurality of classifiers. In some embodiments, the weights of the weighted average are assigned based on the area under the receiver operating characteristic (ROC) curve. In some embodiments, the weights of the weighted average are assigned based on the area under the precision-recall curve. In some embodiments, the weights of the weighted average are assigned based on accuracy. In some embodiments, the weights of the weighted average are assigned based on precision. In some embodiments, the weights of the weighted average are assigned based on recall. In some embodiments, the weights of the weighted average are assigned based on sensitivity.In some embodiments, the weights of the weighted average are assigned based on the F1-score. In some embodiments, the weights of the weighted average are assigned based on specificity. 【0014】 In some aspects, the present specification is to obtain multi-omics data generated from one or more biological fluid samples collected from a subject, where the multi-omics data includes first omics data and second omics data, the first omics data includes a first omics data type including proteomics data, metabolomics data, transcriptomics data, or genomics data, the second omics data includes a second omics data type different from the first omics data type, including proteomics data, metabolomics data, transcriptomics data, or genomics data, obtaining; identifying a subset of first features from the first omics data; identifying a subset of second features from the second omics data; pooling the subsets of the first and second features; and identifying the multi-omics data as indicating or not indicating a disease state based on the pooled subset of features. In some aspects, identifying a subset of first or second features from the first or second omics data includes obtaining univariate data about the features of the first or second omics data and identifying the first or second subset based on the univariate data. In some aspects, the subset of the first or second features is identified from among the features of a classifier for the first or second omics data. In some aspects, identifying a subset of first or second features from the first or second omics data includes obtaining a classifier for the first or second omics data and identifying the first or second subset as the top features of the classifier. In some aspects, identifying a subset of first or second features from the first or second omics data includes obtaining a classifier for the first or second omics data, deleting one or more features from the classifier at a time, and identifying which features, when deleted from the classifier, cause a decrease in the performance of the classifier. 【0015】 In some embodiments, the disease or disorder includes pancreatic cancer. Disclosed herein, in some aspects, is a multi-omics cancer detection method for detecting pancreatic cancer. Disclosed herein, in some aspects, is a method of detecting pancreatic cancer in a subject, comprising: identifying a subject at risk of having pancreatic cancer; obtaining a biological fluid sample from the subject; contacting the biological fluid sample with particles such that the particles adsorb biomolecules including proteins; assaying the biomolecules adsorbed to the particles to generate proteomics data; and classifying the proteomics data as indicative of pancreatic cancer or not indicative of pancreatic cancer. Disclosed herein, in some aspects, is a method of assaying proteins in a biological fluid sample obtained from a subject identified as being at risk of pancreatic cancer to obtain protein measurements; and applying a classifier to the protein measurements to thereby identify the protein measurements as indicative of the subject having pancreatic cancer, wherein the classifier is generated using proteomics data obtained by contacting a training sample with particles such that the particles adsorb proteins in the training sample and assaying the proteins adsorbed to the particles. Disclosed herein, in some aspects, is a treatment method comprising: identifying a tumor in the pancreas of a subject; obtaining a biological fluid sample from the subject; contacting the biological fluid sample with particles such that the particles adsorb biomolecules including proteins; assaying the biomolecules adsorbed to the particles to generate proteomics data; and classifying the proteomics data as indicative of a tumor including pancreatic cancer or not indicative of a tumor including pancreatic cancer.In some aspects, a method for evaluating a subject suspected of having pancreatic cancer, the method comprising measuring a biomarker in a biological fluid sample from the subject, the biomarker comprising A2GL, AKR1B1, ANPEP, ANTXR1, ANTXR2, BTK, CALR, CDH1, CDH11, CDH2, CDHR2, CILP2, CLEC3B, COL18A1, CRP, EXT1, F13A1, FAT1, FGL1, FLT4, ICAM1, IDH2, LCN2, LPP, MAPK1, MAP2K1, MYH9, NOTCH1, NOTCH2, PIGR, PPP2R1A, PRKAR1A, PXDN, RELN, RHOA, S100A8, S100A9, S100A12, SAA1, SAA2, SERPINA3, SLAIN2, SND1, SVEP1, TSP2, TUBB, TUBB1, or VCAN, is disclosed. In some aspects, the present disclosure provides a method comprising assaying a biomolecule in a biological fluid sample obtained from a subject suspected of having pancreatic cancer to obtain a biomolecule measurement value; and applying a classifier to the biomolecule measurement value to identify a protein measurement value as indicating that the subject has pancreatic cancer or as indicating that the subject does not have pancreatic cancer, the classifier being characterized by a receiver operating characteristic (ROC) curve having an area under the curve (AUC) greater than 0.7, greater than 0.75, greater than 0.8, greater than 0.85, greater than 0.9, greater than 0.91, greater than 0.92, greater than 0.93, or greater than 0.94, based on biomolecule measurement features. In some aspects, the AUC is 0.75 or less, 0.8 or less, 0.85 or less, 0.9 or less, 0.91 or less, 0.92 or less, 0.93 or less, 0.94 or less, 0.95 or less, or 0.96 or less. In some embodiments, the biomolecule comprises a protein, a lipid, a metabolite, or a combination thereof. 【0016】 In some embodiments, the disease or disorder includes liver cancer. Disclosed herein, in some aspects, is a multi-omics cancer detection method for detecting liver cancer. Disclosed herein, in some aspects, is a method for detecting liver cancer in a subject, comprising: identifying the subject as being at risk of having liver cancer; obtaining a biofluid sample from the subject; contacting the biofluid sample with particles such that the particles adsorb biomolecules including proteins; assaying the biomolecules adsorbed to the particles to generate proteomics data; and classifying the proteomics data as indicative of liver cancer or not indicative of liver cancer. Disclosed herein, in some aspects, is a method comprising assaying proteins in a biofluid sample obtained from a subject identified as being at risk of liver cancer to obtain protein measurements; and applying a classifier to the protein measurements to thereby identify the protein measurements as indicative of the subject having liver cancer, wherein the classifier is generated using proteomics data obtained by contacting a training sample with particles such that the particles adsorb proteins in the training sample and assaying the proteins adsorbed to the particles. Disclosed herein, in some aspects, is a treatment method comprising: identifying a tumor in the liver of a subject; obtaining a biofluid sample from the subject; contacting the biofluid sample with particles such that the particles adsorb biomolecules including proteins; assaying the biomolecules adsorbed to the particles to generate proteomics data; and classifying the proteomics data as indicative of a tumor including liver cancer or not indicative of liver cancer. Disclosed herein, in some aspects, is a method for detecting liver cancer in a subject, comprising: identifying the subject as being at risk of having liver cancer; obtaining a biofluid sample from the subject; assaying lipids in the biofluid sample to obtain lipid data; and classifying the lipid data as indicative of liver cancer or not indicative of liver cancer. 【0017】 In an embodiment, the disease or disorder includes ovarian cancer. In some aspects herein, a multi-omics cancer detection method for detecting ovarian cancer is disclosed. In some aspects herein, a method for detecting ovarian cancer in a subject, comprising: identifying the subject as being at risk of having ovarian cancer; obtaining a biological fluid sample from the subject; contacting the biological fluid sample with particles such that the particles adsorb a biomolecule containing a protein; assaying the biomolecule adsorbed to the particles to generate proteomics data; and classifying the proteomics data as indicative of ovarian cancer or not indicative of ovarian cancer, is disclosed. In some aspects, identifying the subject as being at risk of having ovarian cancer includes the subject having received a computed tomography (CT) scan indicative of ovarian cancer, having received a magnetic resonance imaging (MRI) scan indicative of ovarian cancer, having received a positron emission tomography (PET) scan indicative of ovarian cancer, having received a transvaginal ultrasound examination indicative of ovarian cancer, having an elevated cancer antigen (CA)-125 level compared to a control or baseline measurement, or having an ovarian cyst, or a combination thereof. In some aspects herein, assaying a protein in a biological fluid sample obtained from a subject identified as being at risk of ovarian cancer to obtain a protein measurement; and applying a classifier to the protein measurement, thereby identifying the protein measurement as indicative of the subject having ovarian cancer, are included, wherein the classifier is generated using proteomics data obtained by contacting a training sample with particles such that the particles adsorb a protein in the training sample and assaying the protein adsorbed to the particles. In some aspects, the protein includes ANTXR2, BMP1, CILP, EIF2AK2, ENO3, F13B, FGL1, or PEBP4.In some embodiments herein, a method of treatment is disclosed that includes identifying a tumor in an ovary of a subject; obtaining a biological fluid sample from the subject; contacting the biological fluid sample with particles such that the particles adsorb a biomolecule containing a protein; assaying the biomolecule adsorbed to the particles to generate proteomics data; and classifying the proteomics data as indicative of a tumor including ovarian cancer or not indicative of ovarian cancer. In some embodiments herein, a method of detecting ovarian cancer in a subject is disclosed that includes identifying the subject as being at risk of having ovarian cancer; obtaining a biological fluid sample from the subject; assaying lipids in the biological fluid sample to obtain lipid data; and classifying the lipid data as indicative of ovarian cancer or not indicative of ovarian cancer. In some embodiments, the lipids include one or more phospholipids. 【0018】 In some embodiments, the disease or disorder includes colorectal cancer. Disclosed herein in some aspects is a multi-omics cancer detection method for detecting colorectal cancer. Disclosed herein in some aspects is a method for detecting colorectal cancer in a subject, comprising: identifying that the subject is at risk of having colorectal cancer; obtaining a biological fluid sample from the subject; contacting the biological fluid sample with particles such that the particles adsorb biomolecules including proteins; assaying the biomolecules adsorbed to the particles to generate proteomics data; and classifying the proteomics data as indicative of colorectal cancer or not indicative of colorectal cancer. Disclosed herein in some aspects is assaying proteins in a biological fluid sample obtained from a subject identified as being at risk of colorectal cancer to obtain protein measurements; and applying a classifier to the protein measurements, thereby identifying the protein measurements as indicative of the subject having colorectal cancer, wherein the classifier is generated using proteomics data obtained by contacting a training sample with particles such that the particles adsorb proteins in the training sample and assaying the proteins adsorbed to the particles. In some aspects, the subject is identified as being at risk of having colorectal cancer by the subject having had a computed tomography (CT) scan indicative of colorectal cancer, having had a liver function test (LFT) indicative of colorectal cancer, having an elevated carcinoembryonic antigen (CEA) level compared to a control or baseline measurement, having blood in the stool, having had a fecal immunochemical test (FIT) indicative of colorectal cancer, or having a colorectal nodule, or a combination thereof. Disclosed herein in some aspects is a treatment method comprising: identifying a tumor in the colon of the subject; obtaining a biological fluid sample from the subject; contacting the biological fluid sample with particles such that the particles adsorb biomolecules including proteins; assaying the biomolecules adsorbed to the particles to generate proteomics data; and classifying the proteomics data as indicative of a tumor including colorectal cancer or not indicative of colorectal cancer. 【0019】In some aspects, this specification includes assaying proteins in a biological fluid sample obtained from a subject identified as having a pulmonary nodule to obtain protein measurement values; applying a classifier to the protein measurement values to evaluate the pulmonary nodule; and (i), (ii), or (iii): (i) the classifier includes protein features of the assayed proteins, and the classifier includes performance characteristics when identifying a pulmonary nodule as cancerous or non-cancerous, and is determined in a dataset derived from a randomized controlled trial of more than 20 subjects with cancerous pulmonary nodules and more than 20 control subjects with non-cancerous pulmonary nodules, and is determined in the dataset without including clinical features in the classifier, and includes performance characteristics including an area under the curve (AUC) of the mean or median receiver operating characteristic (ROC) curve exceeding 0.65 (e.g., exceeding 0.7), (ii) the classifier is generated using proteomics data obtained by contacting a training sample with particles such that the particles adsorb the proteins in the training sample and assaying the proteins adsorbed to the particles, or (iii) assaying the proteins includes contacting the biological fluid sample with particles to adsorb the proteins to the particles and obtaining protein measurement values from the adsorbed proteins. In some aspects, the classifier includes protein features of the assayed proteins, is characterized by a mean ROC curve having a median AUC exceeding 0.7 when identifying a pulmonary nodule as cancerous or non-cancerous, and the AUC exceeding 0.7 is determined in a dataset derived from a randomized controlled trial of more than 20 subjects with cancerous pulmonary nodules and more than 20 control subjects with non-cancerous pulmonary nodules without including non-protein features. In some aspects, the classifier is generated using proteomics data obtained by contacting a training sample with particles such that the particles adsorb the proteins in the training sample and assaying the proteins adsorbed to the particles. In some aspects, assaying the proteins includes contacting the biological fluid sample with particles to adsorb the proteins to the particles and obtaining protein measurement values from the adsorbed proteins.In some aspects, the classifier is trained using deep learning, hierarchical cluster analysis, principal component analysis, partial least squares discriminant analysis, random forest classification analysis, support vector machine analysis, k-nearest neighbor analysis, naive Bayes analysis, K-means clustering analysis, or hidden Markov analysis. In some aspects, evaluating a lung nodule includes identifying protein measurements that indicate that the lung nodule is cancerous. Some aspects include administering a lung cancer treatment to a subject based on the evaluation. In some aspects, the lung cancer treatment includes chemotherapy, radiation therapy, percutaneous ablation, radiofrequency ablation, cryoablation, microwave ablation, chemoembolization, or surgery. In some aspects, the subject is identified as having a lung nodule through the use of a medical imaging device. In some aspects, the classifier identifies lung cancer with a sensitivity and specificity of greater than 60%. In some aspects, the particles include nanoparticles. In some aspects, the particles include lipid particles, metal particles, silica particles, or polymer particles. In some aspects, the particles include a group of physiochemically different nanoparticles. In some aspects, the biological fluid sample includes a blood, serum, or plasma sample. In some aspects, the subject is human. In some aspects, the protein measurements include measurements of proteins selected from the group consisting of APP, IGHG2, SERPING1, SAA2, SERPINF2, GC, IGHA1, HPR, SERPINA3, IGHA1, LTF, SERPINA1, PCSK6, PROS1, BPIF1, C6, CP, A2M, and IGFBP2.In some aspects, this specification involves obtaining protein measurement values by assaying proteins in a blood, serum, or plasma sample by mass spectrometry to obtain protein measurement values, where the sample is obtained from a human subject identified as having a lung nodule using a medical imaging device; applying a classifier to the protein measurement values to evaluate the lung nodule; and selecting or administering a lung cancer treatment method to the subject based on the evaluation; and (i), (ii), or (iii): (i) the classifier includes protein features of the assayed proteins, and the classifier includes performance characteristics when identifying a lung nodule as cancerous or non-cancerous, including a performance characteristic including an area under the curve (AUC) of a receiver operating characteristic (ROC) curve with a median greater than 0.7 determined in a holdout dataset derived from a randomized controlled trial of more than 25 subjects with cancerous lung nodules and more than 25 control subjects with non-cancerous lung nodules, determined using only the protein features of the classifier; (ii) the classifier is generated using proteomics data obtained by contacting a training sample with nanoparticles such that the nanoparticles adsorb the proteins in the training sample and assaying the proteins adsorbed to the nanoparticles; or (iii) assaying the proteins includes contacting a blood, serum, or plasma sample with nanoparticles to adsorb the proteins to the nanoparticles and obtaining protein measurement values from the adsorbed proteins. A method including these is disclosed. 【0020】 In some embodiments, the classifier includes protein features of the assayed protein and is characterized by an average ROC curve having a median AUC exceeding 0.7 when identifying pulmonary nodules as cancerous or non-cancerous, where the AUC exceeding 0.7 is determined without including non-protein features in a holdout dataset derived from a randomized controlled trial of more than 25 subjects with cancerous pulmonary nodules and more than 25 control subjects with non-cancerous pulmonary nodules. In some embodiments, the classifier is generated using proteomics data obtained by contacting a training sample with nanoparticles such that the nanoparticles adsorb the protein in the training sample and assaying the protein adsorbed to the nanoparticles. In some embodiments, assaying the protein includes contacting a blood, serum, or plasma sample with nanoparticles to adsorb the protein to the nanoparticles and obtaining protein measurements from the adsorbed protein. 【0021】 Disclosed herein are methods that, in some aspects, include assaying proteins in a biological fluid sample obtained from a subject identified as having a pulmonary nodule to obtain protein measurements; and applying a classifier to the protein measurements to identify the protein measurements as indicative of whether the pulmonary nodule is cancerous or non-cancerous, where the classifier is characterized by a receiver operating characteristic (ROC) curve having an area under the curve (AUC) exceeding 0.7 based on protein measurement features. In some aspects, the AUC exceeding 0.7 is generated without including non-protein clinical features. In some aspects, the non-protein clinical features include clinical indicators of lung cancer. In some aspects, the proteins include APP, IGHG2, SERPING1, SAA2, SERPINF2, GC, IGHA1, HPR, SERPINA3, IGHA1, LTF, SERPINA1, PCSK6, PROS1, BPIF1, C6, CP, A2M, or IGFBP2. 【0022】 In some aspects, the present specification discloses a method including assaying proteins in a biological fluid sample obtained from a subject having or suspected of having a pulmonary nodule to obtain protein measurement values; and applying a classifier to the protein measurement values to evaluate the pulmonary nodule, wherein the classifier is generated using proteomics data obtained by concentrating the proteins with affinity reagents. In some aspects, the present specification discloses a method including assaying proteins in a biological fluid sample obtained from a subject having or suspected of having a pulmonary nodule to obtain protein measurement values; and applying a classifier to the protein measurement values to thereby identify the protein measurement values as indicating that the pulmonary nodule is cancerous or non-cancerous, wherein the classifier is generated using proteomics data obtained by contacting a training sample with particles such that the particles adsorb the proteins in the training sample and assaying the proteins adsorbed to the particles. Some aspects include obtaining or receiving a biological fluid sample of the subject. In some aspects, the subject is identified as having a pulmonary nodule by medical imaging. In some aspects, the medical imaging includes a computed tomography (CT) scan. Some aspects include performing medical imaging. Some aspects include identifying a pulmonary nodule by medical imaging. Some aspects include generating a report based on the identification of protein measurement values indicating whether the pulmonary nodule is cancerous or non-cancerous. In some aspects, the report includes the likelihood or indication that the pulmonary nodule is cancerous or non-cancerous. Some aspects include outputting or transmitting the report. In some aspects, the report is used by a medical professional in making a diagnosis, providing medical advice, or offering a treatment for the pulmonary nodule. Some aspects include performing a biopsy on the pulmonary nodule if the protein measurement values are classified as indicating that the pulmonary nodule is cancerous. In some aspects, the biopsy confirms the likelihood that the pulmonary nodule is cancerous or non-cancerous. In some aspects, the pulmonary nodule is cancerous.In some embodiments, the lung nodules include non-small cell lung cancer (NSCLC). In some embodiments, the classifier includes features indicative of protein measurements that indicate whether the lung nodule is cancerous or non-cancerous. In some embodiments, the features include control protein measurements, mass spectra, m / z ratios, chromatography results, immunoassay results, or light or fluorescence intensities. In some embodiments, the classifier is trained using deep learning, hierarchical cluster analysis, principal component analysis, partial least squares discriminant analysis, random forest classification analysis, support vector machine analysis, k-nearest neighbor analysis, naive Bayes analysis, K-means clustering analysis, or hidden Markov analysis. In some embodiments, the classifier can identify lung cancer with a sensitivity of 50% or more, 60% or more, 70% or more, 80% or more, or 90% or more. In some embodiments, the classifier can identify lung cancer with a specificity of 50% or more, 60% or more, 70% or more, 80% or more, or 90% or more. Some embodiments include recommending lung cancer treatment for a subject when the protein measurements are classified as indicating that the lung nodule is cancerous. Some embodiments include performing lung cancer treatment on a subject when the protein measurements are classified as indicating that the lung nodule is cancerous. In some embodiments, the lung cancer treatment includes chemotherapy, radiation therapy, percutaneous ablation, radiofrequency ablation, cryoablation, microwave ablation, chemoembolization, or surgery. In some embodiments, the lung nodules are non-cancerous. Some embodiments include observing the subject without performing a biopsy when the protein measurements are classified as indicating that the lung nodule is non-cancerous. In some embodiments, observing the subject without performing a biopsy includes assaying proteins in a second biological fluid sample obtained later from the subject. Some embodiments include assaying proteins in a second biological fluid sample obtained later from the subject. In some embodiments, the particles include nanoparticles. In some embodiments, the particles include lipid particles, metal particles, silica particles, or polymer particles.In some embodiments, the particles comprise carboxylate particles, polyacrylic acid particles, dextran particles, polystyrene particles, dimethylamine particles, amino particles, silica particles, or N-(3-trimethoxysilylpropyl)diethylenetriamine particles. In some embodiments, the particles comprise physiochemically distinct groups of nanoparticles. In some embodiments, assaying a protein comprises contacting a biological fluid sample with the particles such that the particles adsorb the protein to the particles. In some embodiments, assaying a protein comprises measuring a readout indicative of the presence, absence, or amount of a biomolecule. In some embodiments, assaying a protein comprises performing mass spectrometry, chromatography, liquid chromatography, high performance liquid chromatography, solid phase chromatography, lateral flow assay, immunoassay, enzyme-linked immunosorbent assay, western blot, dot blot, or immunostaining, or a combination thereof. In some embodiments, assaying a protein comprises performing mass spectrometry. In some embodiments, the protein comprises a secreted protein. In some embodiments, the biological fluid comprises blood, plasma, or serum. In some embodiments, the pulmonary nodule is less than 3 cm in diameter. In some embodiments, the subject has a plurality of pulmonary nodules. In some embodiments, the subject is a mammal. In some embodiments, the subject is a human. 【0023】 In some aspects, the present specification discloses a method including: obtaining a biological fluid sample from a subject having a pulmonary nodule; contacting the biological fluid sample with particles such that the particles adsorb biomolecules including proteins; assaying the biomolecules adsorbed to the particles to generate proteomics data; and classifying the proteomics data as indicative of whether the pulmonary nodule is cancerous or non-cancerous. In some aspects, the subject is identified as having a pulmonary nodule by medical imaging. In some aspects, the medical imaging includes a computed tomography (CT) scan. Some aspects include performing medical imaging. Some aspects include identifying a pulmonary nodule by medical imaging. Some aspects include performing a biopsy on the pulmonary nodule when the proteomics data is classified as indicative of the pulmonary nodule being cancerous. In some aspects, the biopsy confirms the likelihood that the pulmonary nodule is cancerous or non-cancerous. In some aspects, the pulmonary nodule is cancerous and includes a tumor. In some aspects, the pulmonary nodule includes non-small cell lung cancer (NSCLC). In some aspects, classifying the proteomics data as indicative of whether the pulmonary nodule is cancerous or non-cancerous includes applying a classifier to the proteomics data. In some aspects, the classifier includes features indicative of the likelihood that lung cancer is cancerous or non-cancerous. In some aspects, the classifier is trained using deep learning, hierarchical cluster analysis, principal component analysis, partial least squares discriminant analysis, random forest classification analysis, support vector machine analysis, k-nearest neighbor analysis, naive Bayes analysis, K-means clustering analysis, or hidden Markov analysis. In some aspects, the proteomics data indicates whether the pulmonary nodule is cancerous or non-cancerous with a sensitivity or specificity of about 80% or more. Some aspects include recommending lung cancer treatment for the subject when the proteomics data is classified as indicative of the pulmonary nodule being cancerous. Some aspects include administering lung cancer treatment to the subject when the proteomics data is classified as indicative of the pulmonary nodule being cancerous.In some embodiments, lung cancer treatment includes chemotherapy, radiation therapy, percutaneous ablation, radiofrequency ablation, cryoablation, microwave ablation, chemoembolization, or surgery. In some embodiments, the lung nodule is non-cancerous and benign. Some embodiments include observing a subject without performing a biopsy when proteomic data indicates that the lung nodule is non-cancerous. Some embodiments include monitoring the subject and assaying biomolecules in a second biological fluid sample obtained from the subject at a later time. In some embodiments, the particles include nanoparticles. In some embodiments, the particles include lipid particles, metal particles, silica particles, or polymer particles. In some embodiments, the particles include carboxylate particles, polyacrylic acid particles, dextran particles, polystyrene particles, dimethylamine particles, amino particles, silica particles, or N-(3-trimethoxysilylpropyl)diethylenetriamine particles. In some embodiments, the particles include a group of physiochemically different nanoparticles. In some embodiments, assaying a biomolecule includes measuring a readout indicative of the presence, absence, or amount of the biomolecule. In some embodiments, assaying a biomolecule includes performing mass spectrometry, chromatography, liquid chromatography, high performance liquid chromatography, solid phase chromatography, lateral flow assay, immunoassay, enzyme-linked immunosorbent assay, western blot, dot blot, or immunostaining, or a combination thereof. In some embodiments, assaying a biomolecule includes performing mass spectrometry. In some embodiments, the protein includes a secreted protein. In some embodiments, the biological fluid includes blood, plasma, or serum. In some embodiments, the lung nodule is less than 3 cm in diameter. In some embodiments, the subject has multiple lung nodules. In some embodiments, the subject is a mammal. In some embodiments, the subject is a human. 【0024】 In some aspects, this specification describes a method that includes assaying proteins in a biological fluid sample obtained from a subject suspected of having a pulmonary nodule to obtain protein measurement values; and applying a classifier to the protein measurement values to thereby identify the protein measurement values as indicating that the subject has a pulmonary nodule. The classifier is generated using proteomics data obtained by contacting a training sample with particles such that the particles adsorb the proteins in the training sample and assaying the proteins adsorbed to the particles. Some aspects include recommending that the subject undergo medical imaging, such as a CT scan, if the protein measurement values indicate that the subject has a pulmonary nodule, and not recommending that the subject undergo medical imaging if the protein measurement values do not indicate that the subject has a pulmonary nodule. Some aspects include performing medical imaging, such as a CT scan, on the subject if the protein measurement values indicate that the subject has a pulmonary nodule, and not performing medical imaging on the subject if the protein measurement values do not indicate that the subject has a pulmonary nodule. Some aspects include transmitting or receiving a report regarding medical imaging, such as a CT scan, if the protein measurement values indicate that the subject has a pulmonary nodule, and not transmitting or receiving a report if the protein measurement values do not indicate that the subject has a pulmonary nodule. In some aspects, the protein measurement values indicate that the subject has or is likely to have a pulmonary nodule. In some aspects, the protein measurement values indicate that the subject does not have or is unlikely to have a pulmonary nodule. 【0025】 In some aspects, this specification describes a method that includes assaying proteins in a biological fluid sample obtained from a subject suspected of having lung cancer to obtain protein measurement values; applying a classifier to the protein measurement values to thereby identify the protein measurement values as indicating that the subject has lung cancer, where the classifier is generated using proteomics data obtained by contacting a training sample with particles such that the particles adsorb the proteins in the training sample and assaying the proteins adsorbed to the particles. Some aspects include recommending that the subject undergo medical imaging, such as a CT scan, if the protein measurement values indicate that the subject has lung cancer, and not recommending that the subject undergo medical imaging if the protein measurement values do not indicate that the subject has lung cancer. Some aspects include performing medical imaging, such as a CT scan, on the subject if the protein measurement values indicate that the subject has lung cancer, and not performing medical imaging on the subject if the protein measurement values do not indicate that the subject has lung cancer. Some aspects include transmitting or receiving a report regarding medical imaging, such as a CT scan, if the protein measurement values indicate that the subject has lung cancer, and not transmitting or receiving a report if the protein measurement values do not indicate that the subject has lung cancer. In some aspects, the protein measurement values indicate that the subject has or is likely to have lung cancer. In some aspects, the protein measurement values indicate that the subject does not have or is unlikely to have lung cancer. In some aspects, the lung cancer includes NSCLC. 【0026】 In some aspects, this specification discloses a method including: obtaining a biological fluid sample from a subject suspected of having a pulmonary nodule; contacting the biological fluid sample with particles such that the particles adsorb biomolecules including proteins; assaying the biomolecules adsorbed to the particles to generate proteomics data; and classifying the proteomics data as indicating that the subject has a pulmonary nodule or as not indicating that the subject has a pulmonary nodule based on the proteomics data. Some aspects include recommending that the subject undergo medical imaging such as a CT scan when the proteomics data indicates that the subject has a pulmonary nodule, and not recommending that the subject undergo medical imaging when the proteomics data does not indicate that the subject has a pulmonary nodule. Some aspects include performing medical imaging such as a CT scan on the subject when the proteomics data indicates that the subject has a pulmonary nodule, and not performing medical imaging on the subject when the proteomics data does not indicate that the subject has a pulmonary nodule. Some aspects include transmitting or receiving a report regarding medical imaging such as a CT scan when the proteomics data indicates that the subject has a pulmonary nodule, and not transmitting or receiving a report when the proteomics data does not indicate that the subject has a pulmonary nodule. In some aspects, the proteomics data indicates that the subject has or is likely to have a pulmonary nodule. In some aspects, the proteomics data indicates that the subject does not have or is less likely to have a pulmonary nodule. 【0027】 In some aspects, this specification discloses a method including: obtaining a biological fluid sample from a subject suspected of having lung cancer; contacting the biological fluid sample with particles such that the particles adsorb a biomolecule containing a protein; assaying the biomolecule adsorbed to the particles to generate proteomics data; and classifying the proteomics data, based on the proteomics data, as indicating that the subject has lung cancer or as not indicating that the subject has lung cancer. Some aspects include recommending that the subject undergo medical imaging, such as a CT scan, if the proteomics data indicates that the subject has lung cancer, and not recommending that the subject undergo medical imaging if the proteomics data does not indicate that the subject has lung cancer. Some aspects include performing medical imaging, such as a CT scan, on the subject if the proteomics data indicates that the subject has lung cancer, and not performing medical imaging on the subject if the proteomics data does not indicate that the subject has lung cancer. Some aspects include transmitting or receiving a report regarding medical imaging, such as a CT scan, if the proteomics data indicates that the subject has lung cancer, and not transmitting or receiving a report if the proteomics data does not indicate that the subject has lung cancer. In some aspects, the proteomics data indicates that the subject has or is likely to have lung cancer. In some aspects, the proteomics data indicates that the subject does not have or is unlikely to have lung cancer. 【0028】 In some aspects, this specification discloses a monitoring method that includes obtaining a biological fluid sample from a subject at risk of lung cancer recurrence; contacting the biological fluid sample with particles such that the particles adsorb a biomolecule containing a protein; assaying the biomolecule adsorbed to the particles to generate proteomics data; and classifying the proteomics data based on the proteomics data as indicating that the subject has lung cancer recurrence or as not indicating that the subject has lung cancer recurrence. Some aspects include recommending that the subject undergo medical imaging, such as a CT scan, when the protein measurement indicates that the subject has lung cancer recurrence, and not recommending that the subject undergo medical imaging when the protein measurement does not indicate that the subject has lung cancer recurrence. Some aspects include performing medical imaging, such as a CT scan, on the subject when the protein measurement indicates that the subject has lung cancer recurrence, and not performing medical imaging on the subject when the protein measurement does not indicate that the subject has lung cancer recurrence. Some aspects include transmitting or receiving a report regarding medical imaging, such as a CT scan, when the protein measurement indicates that the subject has lung cancer recurrence, and not transmitting or receiving a report when the protein measurement does not indicate that the subject has lung cancer recurrence. In some aspects, the protein measurement indicates that the subject has or is likely to have lung cancer recurrence. In some aspects, the protein measurement indicates that the subject does not have or is unlikely to have lung cancer recurrence. In some aspects, the subject has received treatment for lung cancer. In some aspects, the lung cancer treatment includes chemotherapy, radiation therapy, or surgery. In some aspects, the cancer is potentially resectable. In some aspects, the lung cancer includes NSCLC. 【0029】 Incorporation by reference All publications, patents, and patent applications referred to in this specification are incorporated by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference herein. To the extent that the publications and patents or patent applications incorporated by reference conflict with the disclosure contained herein, this specification is intended to supersede and / or take precedence over any such conflicting material. 【Brief Description of the Drawings】 【0030】 【Fig. 1A】 Shows a multi-omics approach. 【0031】 【Fig. 1B】 Shows the combination of datasets in a multi-omics approach. 【0032】 【Fig. 2A】 Shows an example of a method for generating and applying a classifier described herein. 【0033】 【Fig. 2B】 A flowchart showing some aspects that can be used in the methods of this specification. 【0034】 【Fig. 3A】 Shows an example of stages in the screening and treatment of patients having or suspected of having a disease state. 【0035】 【Fig. 3B】 Shows an example of stages in the screening and treatment of pancreatic cancer patients. 【0036】 【Fig. 3C】 Shows an example of stages in the screening and treatment of liver cancer patients. 【0037】 【Fig. 4】Shows non-limiting examples of computing devices (in this case, devices with one or more processors, memory, storage, and network interfaces). 【0038】 【Fig. 5】 Shows diagrams of classifiers and feature information according to some aspects described herein. 【0039】 【Fig. 6】 Shows a graph explaining the differential expression of some proteins that can be used to generate a classifier for diagnosing a disease state. 【0040】 【Fig. 7】 Shows a diagram showing the expression of some proteins in samples from diseased subjects compared to control subjects. Some genes were differentially expressed (underexpressed or overexpressed) between groups (NSCLC samples and healthy samples). 【0041】 【Fig. 8】 Shows a scatter plot pair with predictions plotted against each other. RNASeq: Prediction probability (affected) based on RNA-Seq data. Proteomics: Prediction probability (affected) based on proteomics data. RNA_Prot: Prediction probability (affected) based on both RNA-Seq and proteomics data. 【0042】 【Fig. 9】 Includes a receiver operating characteristic (ROC) curve and shows an increase in the area under the curve (AUC) when combining mRNA transcriptomics data and proteomics data compared to either mRNA transcriptomics data or proteomics data alone. 【0043】 【Fig. 10A】Shows the additive multi-omics classification of 30 samples from subjects with a disease state and 30 samples from control subjects, including mRNA transcriptomics data, proteomics data, and combinations of mRNA transcriptomics and proteomics data. 【0044】 【Fig. 10B】 Shows differential mRNAs and proteins whose abundances were measured in the biological fluid samples and used to generate the classifier. 【0045】 【Fig. 11A】 Shows analyses based on proteomics data and microRNA data. The upper panel shows the results of a classifier trained with proteomics data only, the middle panel shows the results of a classifier trained with microRNA data only, and the lower panel shows the results of combining the two data types. 【0046】 【Fig. 11B】 Shows the differentially expressed microRNAs used to generate the classifier. 【0047】 【Fig. 12】 Shows an analysis comparing combinations of three omics data types (proteomics, mRNA, and miRNA) against using only one of each of the three data types. 【0048】 【Fig. 13A】 Shows some aspects that can be used for integrated model classification. 【0049】 【Fig. 13B】 Shows some aspects that can be used for transformation-based classification. 【0050】 【Fig. 14】 Shows the graphical results of the integrated model classification analysis. 【0051】 【Fig. 15】 Shows some aspects of the conversion-based classification analysis. 【0052】 【Fig. 16】 Shows the graph results of the integrated model classification analysis and the conversion-based classification. 【0053】 【Fig. 17】 Shows a non-limiting example of a flowchart of a machine training algorithm for improving the sensitivity and specificity of a classifier for predicting diseases described in this specification. 【0054】 【Fig. 18A】 Shows the ROC curves of some protein data and protein + lipid combination data for disease state classification. 【0055】 【Fig. 18B】 Includes aspects of the sensitivity of the analysis of protein data, lipid data, and protein + lipid combination data for disease state classification. 【0056】 【Fig. 19】 Shows aspects of a two-stage machine learning framework for analyzing and training multiple data types. 【0057】 【Fig. 20A】 Includes aspects of the sensitivity of the analysis of protein data, lipid data, and protein + lipid combination data for disease state classification. 【0058】 【Fig. 20B】 Includes aspects of the sensitivity of the analysis of protein data, lipid data, and protein + lipid combination data for disease state classification. 【0059】 【Fig. 20C】 Shows the ROC curves of some protein data, lipid data and protein + lipid combination data for disease state classification. 【0060】 【Fig. 21】 Shows the ROC curves of several protein data for disease state classification and combined data of protein + lipid + clinical parameter data. 【0061】 【Fig. 22A】 Shows information on several protein data. 【0062】 【Fig. 22B】 Shows the performance aspects of several classifiers. 【0063】 【Fig. 22C】 Shows the performance aspects of several classifiers with and without including several features. 【0064】 【Fig. 23】 Shows aspects of several genetic or transcript data, such as measurement indicators or types, sample types, quality control aspects, or sequencing depths that can be used. 【0065】 【Fig. 24】 Shows various aspects that can be used in several methods described herein. 【0066】 【Fig. 25】 Includes several aspects such as subjects or test results that can be included in the methods described herein. 【0067】 【Fig. 26A】 Includes a table showing explanations of several proteins, OT scores, and several features in a protein classifier. 【0068】 【Fig. 26B】 Includes a table showing explanations of several proteins, OT scores, and several features in a protein classifier. 【0069】 【Fig. 27】 It includes a chart showing the feature importance scores of the lipid classifier. 【0070】 【Fig. 28A】 It shows the results of the Wilcox test for age comparison and the Fisher's exact probability test for gender ratio. 【0071】 【Fig. 28B】 It shows the results of the Wilcox test for age comparison and the Fisher's exact probability test for gender ratio. 【0072】 【Fig. 29A】 It shows the number of proteins detected across all target samples in the analysis of biological fluid samples from control patients and cancer patients. 【0073】 【Fig. 29B】 It shows the number of proteins detected across all target samples in the analysis of biological fluid samples from control patients and cancer patients. 【0074】 【Fig. 30A】 It shows a plot of some top proteins differentially detected in biological fluid samples from cancer patients compared to those from control patients. 【0075】 【Fig. 30B】 It is a plot showing the distribution of OpenTargets (OT) scores. The OT scores (0 - 0.8) are on the x-axis, and the density (0 - 15) is included on the y-axis. 【0076】 【Fig. 31A】 It includes a plot showing the comparison of the total signal median for each sample, type of analyte, and class. 【0077】 【Fig. 31B】 shows box plots of the most significantly different analytes for each omics workflow according to one embodiment; upper left: lipids; lower left: metabolites; and right: proteins). 【0078】 【Fig. 31C】 Shows an example of the performance of a multi-omics classifier that combines proteomics measurements, lipidomics measurements, and metabolomics measurements. 【0079】 【Fig. 32A】 Includes a volcano plot of the intensity differences and P-values of proteins adsorbed to nanoparticles and detected in biological fluid samples from cancer patients compared to biological fluid samples from control patients. In the volcano plot, the magnitude of the difference is shown on the x-axis and the significance is shown on the y-axis, and the most significant analytes are highlighted. 【0080】 【Fig. 32B】 Includes data for the top protein P35442 after a particle-based measurement method. 【0081】 【Fig. 32C】 Includes a volcano plot of the intensity differences and P-values of proteins detected in biological fluid samples from cancer patients compared to biological fluid samples from control patients. In the volcano plot, the magnitude of the difference is shown on the x-axis and the significance is shown on the y-axis, and the most significant analytes are highlighted. 【0082】 【Fig. 32D】 Includes data for the top protein P01011 after proteomics measurements. 【0083】 【Fig. 33A】 Includes a volcano plot of the intensity differences and P-values of lipids detected in biological fluid samples from cancer patients compared to biological fluid samples from control patients. In the volcano plot, the magnitude of the difference is shown on the x-axis and the significance is shown on the y-axis, and the most significant analytes are highlighted. 【0084】 【Fig. 33B】 It includes data of the top lipid CER (d18:1_18:0) after lipidomics measurement. 【0085】 【Fig. 34A】 It includes a volcano plot of the intensity difference and P-value of metabolites detected in biological fluid samples from cancer patients compared with those from control patients. In the volcano plot, the magnitude of the difference is shown on the x-axis, the significance is shown on the y-axis, and the most significant analytes are highlighted. 【0086】 【Fig. 34B】 It includes data of the top metabolite AICAR after metabolomics measurement. 【0087】 【Fig. 35A】 It shows the classification of cancer and healthy samples by UMAP projection based on combined data. 【0088】 【Fig. 35B】 It shows the classification of cancer and healthy samples by PCA projection based on combined data. 【0089】 【Fig. 35C】 It shows the classification of cancer and healthy samples by UMAP projection based on Proteograph data. 【0090】 【Fig. 35D】 It shows the classification of cancer and healthy samples by PCA projection based on Proteograph data. 【0091】 【Fig. 35E】 It shows the classification of cancer and healthy samples by UMAP projection based on PiQuant data. 【0092】 【Fig. 35F】 It shows the classification of cancer and healthy samples by PCA projection based on PiQuant data. 【0093】 【Fig. 35G】 Classification of cancer and healthy samples by UMAP projection based on lipid data is shown. 【0094】 【Fig. 35H】 Classification of cancer and healthy samples by PCA projection based on lipid data is shown. 【0095】 【Fig. 35I】 Classification of cancer and healthy samples by UMAP projection based on metabolite data is shown. 【0096】 【Fig. 35J】 Classification of cancer and healthy samples by PCA projection based on metabolite data is shown. 【0097】 【Fig. 36】 Features of proteins, lipids, and metabolites included in the classifier are shown. 【0098】 【Fig. 37】 Performance of the classifier in multi - omics studies is shown, including the receiver operating characteristic (ROC) curve for disease state classification. The area under the curve (AUC) values are also included in the figure, and the 90% confidence intervals are shown in parentheses. 【0099】 【Fig. 38A】 Performance of the classifier trained using data from genomic assays is shown, including the ROC curve for disease state classification. The AUC values at the bottom of the figure are shown as ± values based on 90% confidence. 【0100】 【Fig. 38B】Shows the performance of classifiers trained with data from genomics assays ("Genomics"), classifiers trained with data from mass spectrometry assays ("Mass-spec"), and classifiers trained with data from genomics and mass spectrometry assays ("Combined"). The data shown in the figure includes ROC curves for disease state classification. The AUC values include ± values based on 90% confidence. 【0101】 【Fig. 39A】 Shows a graphical summary of 18 samples from liver cancer subjects used in Example 17. 【0102】 【Fig. 39B】 Shows the coefficient of variation (CV) values of several peptides and proteins obtained in the studies described herein. 【0103】 【Fig. 39C】 Shows an exemplary protein abundance heatmap of samples from subjects with liver cancer and healthy subjects. 【0104】 【Fig. 39D】 Shows examples of differences in protein abundances identified in samples from subjects with liver cancer or healthy subjects after contacting the samples with various particles described herein. 【0105】 【Fig. 39E】 Includes a graph showing that lipidomics data obtained from samples is highly reproducible. 【0106】 【Fig. 39F】 Shows that samples from subjects with liver cancer showed different lipid profiles and healthy controls. The top 50 lipids based on the p-value of this analysis are shown for each patient sample. 【0107】 【Fig. 39G】Shows the univariate lipid differences in samples from subjects with liver cancer compared to healthy subjects. 【0108】 【Fig. 40A】 Shows a graph summary of nine samples from ovarian cancer subjects used in Example 19. 【0109】 【Fig. 40B】 Shows an exemplary protein abundance heatmap of samples from subjects with ovarian cancer and healthy subjects. 【0110】 【Fig. 40C】 Shows the univariate lipid differences in samples from subjects with ovarian cancer compared to healthy subjects. 【0111】 【Fig. 41】 Shows examples of stages in the screening and treatment of colorectal cancer patients. 【0112】 【Fig. 42】 Shows the breakdown of age and gender of 268 subjects in the NSCLC biomarker discovery study. 【0113】 【Fig. 43】 Shows the number of proteins for each study group including healthy, comorbidities, NSCLC stage 1 "NSCLC_1", NSCLC stage 2 "NSCLC_2", NSCLC stage 3 "NSCLC_3", and NSCLC stage 4 "NSCLC_4". 【0114】 【Fig. 44】 Shows the number of proteins in depleted plasma DP and particle panel. 【0115】 【Fig. 45】 Shows an overview of the fractional detection of proteins across all subjects for 10 particle types in the particle panel and depleted plasma (DP) against the average abundance of said proteins. 【0116】 【Fig. 46】 This shows the performance of a cross-validated particle panel classifier, where the x-axis represents the percentage of false positive classifications and the y-axis represents the percentage of true positive classifications. 【0117】 【Fig. 47】 Graphs of the random forest model for healthy vs NSCLC (stages 1, 2, and 3) for depleted plasma (left) and 10-particle panel (right), with the false positive rate on the x-axis and the true positive rate on the y-axis. 【0118】 【Fig. 48】 This shows the performance of classifier features across the study sample. 【0119】 【Fig. 49】 This shows the results of repeating 10 rounds of 10-fold cross-validation with randomized subject class assignments using the false positive rate on the x-axis and the true positive rate on the y-axis 10 times. 【0120】 【Fig. 50】 This shows the ROC plots for 13 peptides by MRM-MS and 2 proteins by ELISA after removing the proteins found in depleted plasma. 【0121】 【Fig. 51】 This shows the random forest model for all study group comparisons. 【0122】 【Fig. 52】 This shows some identified features in study group comparisons. 【0123】 【Fig. 53】 This shows the number of proteins (e.g., the number of proteins identified from corona analysis) for panel sizes ranging from 1 particle type to 12 particle types. 【0124】 【Fig. 54】 This shows examples of biomarkers. 【0125】 【Fig. 55】 A diagram showing a non - limiting example of a web / mobile application providing system (in this case, a system providing a browser - based and / or native mobile user interface); and 【0126】 【Fig. 56】 A diagram showing a non - limiting example of a cloud - based web / mobile application providing system (in this case, the system includes elastically load - balanced, automatically scaled web server and application server resources and synchronized and replicated databases). 【0127】 【Fig. 57】 The ROC curve of the pulmonary nodule classifier is shown, and the sensitivity and corresponding specificity are listed. 【0128】 【Fig. 58】 Shows the feature information and importance of the pulmonary nodule classifier shown in FIG. 57. 【0129】 【Fig. 59】 Shows some aspects of the samples used in the studies described herein. 【0130】 【0131】 【Fig. 60】 Shows the number of protein groups observed in the process control samples. 【0132】 【Fig. 61】 Shows some coefficient of variation (CV) values. 【0133】 【Fig. 62】 Includes a protein abundance heatmap of samples from subjects with malignant and benign pulmonary nodules. 【0134】 【Fig. 63】 It includes a volcano plot plotting the logarithmic fold change of protein abundance against the negative logarithm of the p-value. 【0135】 【Fig. 64】 Examples of some proteins from the initial univariate analysis are shown. 【0136】 【Fig. 65A】 It includes a graph showing some proteins upregulated in a biological fluid sample from a subject with malignant lung nodules. 【0137】 【Fig. 65B】 It includes a graph showing some proteins downregulated in a biological fluid sample from a subject with malignant lung nodules. 【0138】 【Fig. 66】 It includes a graph showing that differentially expressed proteins were enriched in metabolic pathways and phosphorylation pathways. 【0139】 【Fig. 67】 Some extrapolated mRNA data showing differentially expressed proteins in metabolic pathways are shown. 【0140】 【Fig. 68】 It is an image showing the location where some samples were collected for the study. 【0141】 【Fig. 69A】 Some aspects of the research subjects and proteomics platforms that can be used in the methods described herein are shown. 【0142】 【Fig. 69B】 Some aspects of the proteomics platforms that can be used in the methods described herein are shown. 【0143】 【Fig. 69C】Shows some additional multi-omics aspects. 【0144】 【Fig. 70】 Includes a graphical depiction of the coefficient of variation (CV) values obtained in the studies described herein. 【0145】 【Fig. 71】 Includes the empirical detection power curve of protein changes in the studies described herein. 【0146】 【Fig. 72】 Includes a graphical display of the detected protein groups and the number of peptides obtained in the studies described herein. 【0147】 【Fig. 73】 Includes a graphical display of protein concentration versus the natural logarithm of protein intensity data obtained in the studies described herein. 【0148】 【Fig. 74】 Includes a graphical display of the protein concentration of the data obtained in the studies described herein. 【0149】 【Fig. 75A】 Includes the median of the normalized log intensity CV of the proteins detected in 100% of the samples. 【0150】 【Fig. 75B】 Includes the median of the normalized log intensity CV of the proteins detected in at least 25% of the samples. 【0151】 【Fig. 76】 Includes the number of unique protein groups within some of the sample data. 【0152】 【Fig. 77A】 Includes relative fluorescence units versus concentration for some of the standard curves. 【0153】 【Fig. 77B】 Contains the relative fluorescence units of several standard curves. 【0154】 【Fig. 78A】 Contains the peptide yields of several nanoparticles used in the experiments described herein. 【0155】 【Fig. 78B】 Contains the peptide yields of several nanoparticles used in the experiments described herein. 【0156】 【Fig. 79A】 Contains a graph of MS1 intensity over time. 【0157】 【Fig. 79B】 Contains the intra-day CV of MS1 intensity. 【0158】 【Fig. 80A】 Contains a graph of iRT peptides ranked by FWHM. 【0159】 【Fig. 80B】 Contains a plot showing retention time. 【0160】 【Fig. 81A】 Contains a plot showing the protein group number distribution per sample. 【0161】 【Fig. 81B】 Contains the intra-day CV of MS1 intensity. 【0162】 【Fig. 82】 Contains a volcano plot of the intensity difference and P-value of peptides detected in a biological fluid sample. In the volcano plot, the median difference in peptide-level intensity is displayed on the x-axis, and the harmonic mean-based peptide P-value is displayed on the y-axis. 【0163】 【Fig. 83】Includes graphs showing several transitions for the peptide ANVFVQLPR (SEQ ID NO: 165) from protein P35858 in the benign and malignant groups. 【0164】 【Fig. 84】 Includes a graph showing the comparison of the significant difference between the OpenTarget (OT) score of lung cancer and the peptide. In the graph, the OpenTarget score is displayed on the x-axis and the P-value is displayed on the y-axis. 【0165】 【Fig. 85】 Includes a volcano plot of the intensity difference and P-value for metabolites in lung nodule subjects. In the volcano plot, the median difference in intensity is displayed on the x-axis and the P-value is displayed on the y-axis. 【0166】 【Fig. 86】 Includes a figure showing the seer-based lung discovery sample cohort. The figure shows that among 589 eligible subjects, 186 subjects met all the criteria. 【0167】 【Fig. 87】 Shows a diagram of the stepwise approach to discover the classifier of version 1, the classifier of version 2, and the classifier of version 3 through test development. 【0168】 【Fig. 88】 Includes a graph showing the receiver operating characteristic curve for analyte classes. The graph includes curves for proteins, metabolites, and lipids. 【0169】 【Fig. 89】 Includes a volcano plot of the intensity difference and P-value for peptides in lung nodule subjects. In the volcano plot, the median difference in peptide-level intensity is displayed on the x-axis and the harmonic mean-based peptide p-value is displayed on the y-axis. 【0170】 【Fig. 90】It includes a graph showing several transitions for the peptide LEYLLLSR (SEQ ID NO: 166) from protein P35858 in the benign and malignant groups. 【0171】 【Fig. 91】 It includes a graph showing several transitions for the peptide ANVFVQLPR (SEQ ID NO: 165) from protein P35858 in the benign and malignant groups. 【0172】 【Fig. 92】 It includes a graph showing several transitions for the peptide FLNVLSPR (SEQ ID NO: 167) from protein P17936 in the benign and malignant groups. 【0173】 【Fig. 93】 An image representing StringDB is shown. This image highlights the known interactions of IGFALS and IGFBP3. 【0174】 【Fig. 94】 It includes a volcano plot of the intensity differences and P - values for metabolites in lung nodule subjects. In the volcano plot, the median difference in intensity is shown on the x - axis and the P - value is shown on the y - axis. 【0175】 【Fig. 95】 It includes a graph showing the amounts of biopterin metabolites in the benign and malignant groups. The graph shows the type of study group on the x - axis and the amount of metabolite on the y - axis. 【0176】 【Fig. 96】 It includes a volcano plot of the intensity differences and P - values for lipids in lung nodule subjects. In the volcano plot, the median difference in intensity is shown on the x - axis and the P - value is shown on the y - axis. 【0177】 【Fig. 97】It includes a graph showing the comparison of the significant differences between the OpenTarget (OT) score of lung cancer and peptides. In the graph, the OpenTarget score is shown on the x-axis and the P-value is shown on the y-axis. 【0178】 【Fig. 98】 It shows a diagram of a step-by-step approach to discovering the classifier of version 1, the classifier of version 2, and the classifier of version 3 through test development. 【0179】 【Fig. 99】 It includes graphs of the pre-test probability of subjects with benign nodules, as well as the pre- and post-test probabilities of subjects with benign nodules. The graph shows the probability on the x-axis and the number of subjects on the y-axis. 【0180】 【Fig. 100】 It includes a graph comparing sensitivity and specificity. The graph shows specificity on the x-axis and sensitivity on the y-axis. 【0181】 【Fig. 101】 It shows the ROC curve of 223 subjects with mRNA data in colorectal cancer (CRC) research. The false positive rate is shown on the x-axis and the true positive rate is shown on the y-axis. The AUC value is provided. 【0182】 【Fig. 102】 It includes a volcano plot showing the expression differences of various genes in colorectal cancer research. 【0183】 【Fig. 103】 It shows the ROC curves of ProteoGraph, mRNA, and ProteoGraph + mRNA. The respective AUC values are provided. 【0184】 【Fig. 104】 It shows the ROC curves of ProteoGraph, PiQuant, mRNA, microRNA, and ProteoGraph + PiQuant + mRNA + microRNA. The respective AUC values are provided. 【0185】 【Fig. 105】 The ROC curves of PiQuant, mRNA, and PiQuant + mRNA are shown. The respective AUC values are provided. 【0186】 【Fig. 106】 The ROC curves for classification based on distinct types or combined types of biomolecules are shown. 【0187】 【Fig. 107A】 Figure 107A shows the results of the Wilcox test for age comparison and the Fisher's exact test for gender ratio. 【0188】 【Fig. 107B】 Figure 107B shows the results of the Wilcox test for age comparison and the Fisher's exact test for gender ratio. 【0189】 【Fig. 108A】 Figure 108A shows the number of proteins detected across the entire target samples in the analysis of biological fluid samples from control patients and cancer patients. 【0190】 【Fig. 108B】 Figure 108B shows the number of proteins detected across the entire target samples in the analysis of biological fluid samples from control patients and cancer patients. 【0191】 【Fig. 108C】 Figure 108C shows the reproducibility of the platform, indicating the ability to detect biological signals. Analysis groups: C = control, S = sample. Left panel: Only proteins in the detection / analysis groups with n > 1 were retained. Of the 2,089, two features with CV > 300% were removed for clarity. Right panel: Only proteins in the detection / analysis groups with n > 1 were retained. Of the 7,672, 48 features with CV > 300% were removed for clarity. 【0192】 【Fig. 108D】 Figure 108D shows that more than 5,000 proteins were detected in the feasibility study of 212 subjects. A median of 4 peptides per protein were detected for proteins present in >25% of the samples in a complete UniProt human proteome database containing search parameters: 0.1% peptide / protein FDR, default timsTOF parameters, contaminants (50% reverse decoys). 【0193】 【Fig. 108E】 Figure 108E shows that a large number of proteins were reproducibly detected across samples. Both the identification of complementary proteins and the identification of common proteins were obtained with individual nanoparticles. Distinct protein groups were shown along with classification by sample and collection site for each sample / particle + panel. 【0194】 【Fig. 108F】 Figure 108F shows enhanced proteome coverage for detecting known cancer-related proteins. All matching proteins detected from the samples were plotted on the HPPP curve. GeneCards data used matching gene ids and scores reported from the search term "cancer". The detected HPPP1 proteins covered an 8-digit scale difference: highest concentration: P00450 - ceruloplasmin; 830,000 ng / mL, and lowest concentration: Q7Z627 - E3 ubiquitin-protein ligase HUWE1; 0.0034 ng / mL. 【0195】 【Fig. 108G】 Figure 108G shows deep and efficient large-scale plasma proteomics. 【0196】 【Fig. 108H】 Figure 108H shows the quantitative performance of Proteograph suitable for large-scale studies. 【0197】 【Fig. 108I】 Figure 108I shows the reproducibility of large-scale protein enrichment by Proteograph. The reproducibility of Proteograph enrichment was ideally suitable for biomarker discovery. Data were collected over 191 enrichments of the same sample. The scope of the collection included three instruments, three cohort studies, five operators, an eight-month execution time, 121 plates, and over 1,500 subject samples. 【0198】 【Fig. 108J】 Figure 108J shows the reproducibility of the platform over time (number of months) and instruments. The median MS1 peak area of the iRT peptides was less than 15% for all and mostly less than 10%. 【0199】 【Fig. 108K】 Figure 108K shows the application of the platform to pancreatic cancer biomarker discovery. 【0200】 【Fig. 109A】 Figure 109A shows a plot of some of the top proteins differentially detected in biological fluid samples from cancer patients compared to those from control patients. 【0201】 【Fig. 109B】 Figure 109B is a plot showing the distribution of the OpenTargets (OT) score. The OT score (0 - 0.8) is on the x-axis, and the density (0 - 15) is included on the y-axis. 【0202】 【Fig. 110A】 Figure 110A includes a plot showing the comparison of the median total signal for each sample, type of analyte, and class. 【0203】 【Fig. 110B】 Figure 110B shows box-and-whisker plots of the most significantly different analytes for each omics workflow (A: lipids, B: metabolites, and C: proteins). 【0204】 【Fig. 110C】 FIG. 110C shows the performance of an exemplary multimarker classifier that combines proteomics measurements, lipidomics measurements, and metabolomics measurements. 【0205】 【Fig. 111A】 FIG. 111A includes a volcano plot of the intensity differences and P-values of proteins adsorbed to nanoparticles and detected in a biological fluid sample from a cancer patient compared to a biological fluid sample from a control patient. In the volcano plot, the magnitude of the difference is shown on the x-axis and the significance is shown on the y-axis, and the most significant analytes are highlighted. 【0206】 【Fig. 111B】 FIG. 111B includes data for the top protein P35442 after a particle-based measurement method. 【0207】 【Fig. 111C】 FIG. 111C includes a volcano plot of the intensity differences and P-values of proteins detected in a biological fluid sample from a cancer patient compared to a biological fluid sample from a control patient. In the volcano plot, the magnitude of the difference is shown on the x-axis and the significance is shown on the y-axis, and the most significant analytes are highlighted. 【0208】 【Figure 111D】 FIG. 111D includes data for the top protein P01011 after proteomics measurements. 【0209】 【Figure 112A】 FIG. 112A includes a volcano plot of the intensity differences and P-values of lipids detected in a biological fluid sample from a cancer patient compared to a biological fluid sample from a control patient. In the volcano plot, the magnitude of the difference is shown on the x-axis and the significance is shown on the y-axis, and the most significant analytes are highlighted. 【0210】 【Figure 112B】Figure 112B includes data on the top lipid CER(d18:1_18:0) after lipidomics measurement. 【0211】 【Figure 113A】 Figure 113A includes a volcano plot of the intensity differences and P-values of metabolites detected in biological fluid samples from cancer patients compared to those from control patients. In the volcano plot, the magnitude of the difference is shown on the x-axis and significance is shown on the y-axis, and the most significant analytes are highlighted. 【0212】 【Figure 113B】 Figure 113B includes data on the top metabolite AICAR after metabolomics measurement in biological fluid samples. 【0213】 【Figure 114A】 Figure 114A shows the classification of cancer and healthy controls by UMAP projection based on combined data generated from biological fluid samples. 【0214】 【Figure 114B】 Figure 114B shows the classification of cancer and healthy controls by PCA projection based on combined data generated from biological fluid samples. 【0215】 【Figure 114C】 Figure 114C shows the classification of cancer and healthy controls by UMAP projection based on Proteograph data generated from biological fluid samples. 【0216】 【Figure 114D】 Figure 114D shows the classification of cancer and healthy controls by PCA projection based on Proteograph data generated from biological fluid samples. 【0217】 【Figure 114E】 Figure 114E shows the classification of cancer and healthy controls by UMAP projection based on PiQuant data generated from biological fluid samples. 【0218】 【Figure 114F】 Figure 114F shows the classification of cancer and healthy controls by PCA projection based on PiQuant data generated from a biological fluid sample. 【0219】 【Figure 114G】 Figure 114G shows the classification of cancer and healthy controls by UMAP projection based on lipid data generated from a biological fluid sample. 【0220】 【Figure 114H】 Figure 114H shows the classification of cancer and healthy controls by PCA projection based on lipid data generated from a biological fluid sample. 【0221】 【Figure 114I】 Figure 114I shows the classification of cancer and healthy controls by UMAP projection based on metabolite data generated from a biological fluid sample. 【0222】 【Figure 114J】 Figure 114J shows the classification of cancer and healthy controls by PCA projection based on metabolite data generated from a biological fluid sample. 【0223】 【Figure 115】 Features of proteins, lipids, and metabolites included in the classifier. 【0224】 【Figure 116】 Figure 116 shows the performance of the classifier in a multi-omics study, including the receiver operating characteristic (ROC) curve for disease state classification. The area under the curve (AUC) value is also included in the figure, with the 90% confidence interval shown in parentheses. 【0225】 【Figure 117A】 Figure 117A shows the performance of a classifier trained using data from a genomics assay, including the ROC curve for disease state classification. The AUC values at the bottom of the figure are shown as ± values based on 90% confidence. 【0226】 【Figure 117B】 Figure 117B shows the performance of classifiers trained on data from genomic assays ("Genomics"), classifiers trained on data from mass spectrometry assays ("Mass spectrometry"), and classifiers trained on data from genomic and mass spectrometry assays ("Combination"). The data shown in the figure includes ROC curves for disease state classification. The AUC values include ± values based on 90% confidence. 【0227】 【Figure 118A】 Figure 118A shows a volcano plot indicating the intensity differences between biological fluid samples from subjects with pancreatic cancer and biological fluid samples from healthy controls. 【0228】 【Figure 118B】 Figure 118B shows the study comparison groups (H: healthy, PC: pancreatic cancer). 124 out of 3,381 detected proteins were statistically significant. 【0229】 【Figure 119A】 Figures 119A - C show volcano plots indicating the differential abundance of lipid species between biological fluid samples from subjects with pancreatic cancer and biological fluid samples from healthy controls. 【Figure 119B】 Figures 119A - C show volcano plots indicating the differential abundance of lipid species between biological fluid samples from subjects with pancreatic cancer and biological fluid samples from healthy controls. 【Figure 119C】 Figures 119A - C show volcano plots indicating the differential abundance of lipid species between biological fluid samples from subjects with pancreatic cancer and biological fluid samples from healthy controls. 【0230】 【Figure 120A】 Figure 120A shows the quantitative performance of Proteograph suitable for large - scale studies (e.g., the study of Example 7). 【0231】 【Figure 120B】Figure 120B shows the reproducibility of large-scale protein enrichment by Proteograph. The reproducibility of Proteograph enrichment was ideally suitable for biomarker discovery. The system provides high-throughput, reproducible deep proteome coverage for new discoveries. Quantitative deep untargeted proteomics biomarker studies were made possible by Proteograph reproducibility. Large-scale protein enrichment by Proteograph was highly reproducible (NP1 = 0, NP2 = 0, NP3 = 2, NP4 = 0, and NP5 = 2). 【0232】 【Figure 121A】 Figure 121A shows the evaluation of K562 precursor detection in SWATH vs Zeno SWATH DIA. A minimum 26% increase in precursor identification was detected using Zeno SWATH DIA. All data were generated from the pr and pg matrices from the DIA-NN output (all called and quantified precursors and proteins were identified). All data were searched in DIA-NN with "robust LC" and the SCIEX K562 spectral library. 【0233】 【Figure 121B】 Figure 121B shows the evaluation of K562 precursor detection in SWATH vs Zeno SWATH DIA. A minimum 13% increase in protein group identification was detected using Zeno SWATH DIA. All data were generated from the pr and pg matrices from the DIA-NN output (all called and quantified precursors and proteins were identified). All data were searched in DIA-NN with "robust LC" and the SCIEX K562 spectral library. 【0234】 【Figure 122】 Figure 122 shows improved sensitivity in increasing the number of low-abundance peptide species detected. Detection of low-abundance peptides was improved using Zenon SWATH DI compared to SWATH. 【0235】 【Figure 123】 Figure 123 shows a graph generated from all eligible precursors. The data was searched in DIA-NN using the "Robust LV" and the SCIEX K562 spectral library. 【0236】 【Figure 124】 Figure 124 shows that the quantitative sensitivity increases with mass in SWATH and Zeno SWATH DIA. The MS1 peak areas (K562) of Zeno SWATH DIA were distributed among peptides with low abundances. 【0237】 【Figure 125A】 Figure 125A shows that Zeno SWATCH DIA acquisition yielded higher K562 MS2-based precursor amounts across different peptide injection masses based on all eligible precursors, compared to SWATH acquisition alone. The data was searched in DIA-NN using the "Robust LC" and the SCIEX K562 spectral library. 【0238】 【Figure 125B】 Figure 125B shows that Zeno SWATH DIA acquisition yielded lower CV(5) with respect to the amount of K562 precursor levels across different peptide injection masses based on all quantified precursors, compared to SWATH acquisition alone. The data was searched in DIA-NN using the "Robust LC" and the SCIEX K562 spectral library. 【0239】 【Figure 126】 Figure 126 shows that Zeno Swatch DIA MS / MS acquisition yielded 53 - 85% higher peptide identifications from Proteograph generated from pooled control samples, compared to SWATH MS / MS DIA acquisition. 【0240】 【Figure 127】Figure 127 shows 2,357 protein groups across all five nanoparticles in a representative subject cohort. 1,077 protein groups were identified in at least 25% of patient samples. 【0241】 【Figure 128A】 Figure 128A shows that numerous proteins were reproducibly detected across samples. Identification of both complementary and common proteins was obtained by individual nanoparticles. 【0242】 【Figure 128B】 Figure 126B shows improved sensitivity corresponding to the detection of low-abundance peptides in Proteograph peptide detection. 【0243】 【Figure 129】 Figure 129 shows the distribution of patients used in the lung nodule evaluation study. 【0244】 【Figure 130A】 Figure 130A shows the evaluation of lung nodules. Univariate analysis was performed for each omics using the Wilcoxon test and the Benjamini-Hochberg multiple hypothesis testing correction procedure. 672 lipids, 376 metabolites, 557 miRNAs, 131,603 mRNA transcripts, 555 peptides (targeted), and 9,861 peptide-NPs (untargeted) were detected. No univariate molecular features were statistically significant after correction by multiple hypothesis testing in the analysis across omics. 【0245】 【Figure 130B】 Figure 130B shows lung nodule classifiers trained separately in each omics, with an AUC of 0.62 obtained from untargeted proteomics data. 【0246】 【Figure 130C】Figure 130C shows that by training the classifier by combining all omics, an AUC of 0.6 was obtained. The joint model was trained using the individual omics with the best performance, for example, untargeted proteomics and mRNA. This classifier also had an AUC of 0.6. Adding clinical covariates related to the Mayo score to these classifiers did not improve the model performance. 【0247】 【Figure 131】 Figure 131 shows the development of a high-risk lung cancer screening classifier. By comparing benign vs. malignant nodule classes, features that are likely to be cancer-specific were identified. By training a high-risk vs. malignant classifier with the features selected by the benign vs. malignant comparison, the model was more likely to identify cancer-specific signals rather than differences arising from confounding covariates. Careful confounding factor analysis was required for the trained classifier. 【0248】 【Figure 132A】 Figure 132A shows a volcano plot for the malignant vs. high-risk comparison. 725 lipids, 371 metabolites, 480 miRNAs, 111,949 mRNA transcripts, and 509 peptides (targeted) were detected. The light gray dots identified features that were significantly different after Benjamini-Hochberg multiple hypothesis testing correction. These features represented a mix of cancer-specific and non-specific differences between the groups. 【0249】 【Figure 132B】 Figure 132B shows a volcano plot for the malignant vs. high-risk comparison. The light gray dots identified features that were significantly different (without multiple hypothesis testing correction) in the malignant vs. benign comparison. These identified features representing cancer-specific signals. By training the classifier with the partially selected cancer-specific features, the effect of confounding factors in the malignant vs. high-risk classification could be avoided. 【0250】 【Figure 133】 Figure 133 shows the first classifier trained using features partially selected from the benign vs. malignant lung nodule comparison that exhibits good performance for the comparison of malignant vs. high-risk. 831 filtered features including a combination of mRNA, lipids, metabolites, peptides, and miRNAs were used for training. 【0251】 【Figure 134】 Figure 134 shows proteomic analysis in biological fluid samples of subjects with pancreatic cancer and control subjects. 【0252】 【Figure 135】 Figure 135 shows potential multi-omics pancreatic cancer molecular biomarkers spanning from genotype to phenotype. Molecular features that were differentially present and statistically significant in cancer vs. control were identified by the light gray dots. 【0253】 【Figure 136】 Figure 136 includes an ROC plot showing improvement in classifier performance when combining features from different data types. 【0254】 【Figure 137】 Figure 137 includes plots showing the first cross-factor analysis. 【0255】 【Figure 138A】 Figure 138A shows classifier data from PDAC patients and controls based on metabolomics. 【0256】 【Figure 138B】 Figure 138B shows classifier data from PDAC patients and controls based on Proteograph. 【0257】 【Figure 138C】 Figure 138C shows classifier data from PDAC patients and controls based on PiQuant. 【0258】 【Figure 138D】 Figure 138D shows classifier data from PDAC patients and controls based on lipidomics. 【0259】 【Figure 138E】 Figure 138E shows classifier data from PDAC patients and controls based on RNA. 【0260】 【Figure 138F】 Figure 138F shows classifier data from PDAC patients and controls based on copy number variations. 【0261】 【Figure 138G】 Figure 138G shows classifier data from PDAC patients and controls based on fragmentomics. 【0262】 【Figure 138H】 Figure 138H shows the levels of CA-19-9 in the biological fluids of subjects with pancreatic cancer at various stages and control subjects. 【0263】 【Figure 138I】 Figure 138I shows classifier data from PDAC patients and controls based on carbohydrate antigen 19-9 (CA-19-9), alone or in combination with PiQuant data. 【0264】 【Figure 139】 Figure 139 shows some features useful for training to generate a classifier or applying the classifier to a subject suspected of having cancer such as pancreatic cancer. 【0265】 【Figure 140A】 Figure 140A shows some copy number variation features used in a classifier for evaluating biological fluid samples from subjects with pancreatic cancer or subjects without cancer. 【0266】 【Figure 140B】Figure 140B shows several mRNA features used in a classifier for evaluating biological fluid samples from subjects with pancreatic cancer or subjects without cancer. 【0267】 【Figure 140C】 Figure 140C shows several microRNA features used in a classifier for evaluating biological fluid samples from subjects with pancreatic cancer or subjects without cancer. 【0268】 【Figure 140D】 Figure 140D shows several protein features used in a classifier for evaluating biological fluid samples from subjects with pancreatic cancer or subjects without cancer. Here, the protein features included measurements obtained using an internal standard. UniProt ID numbers were included for each feature. 【0269】 【Figure 140E】 Figure 140E shows several peptide features used in a classifier for evaluating biological fluid samples from subjects with pancreatic cancer or subjects without cancer. Here, the peptide features included measurements obtained using an internal nanoparticle set (NP1, NP2, NP3, NP4, or NP5). Amino acid sequences and nanoparticle designations were included for each feature. 【0270】 【Figure 140F】 Figure 140F shows several protein features used in a classifier for evaluating biological fluid samples from subjects with pancreatic cancer or subjects without cancer. Here, the protein features included measurements obtained using an internal nanoparticle set (NP1, NP2, NP3, NP4, or NP5). UniProt ID numbers and nanoparticle designations were included for each feature. 【0271】 【Figure 140G】FIG. 140G shows some lipid features used in a classifier for evaluating biological fluid samples from subjects with pancreatic cancer or subjects without cancer. 【0272】 【Figure 140H】 FIG. 140H shows some metabolite features used in a classifier for evaluating biological fluid samples from subjects with pancreatic cancer or subjects without cancer. 【0273】 【Figure 141A】 FIG. 141A shows an ROC curve for a machine learning model that includes a combination of PiQuant, metabolomics, lipidomics, and CA-19-9. The respective AUC values are provided. 【0274】 【Figure 141B】 FIG. 141B shows an ROC curve for a machine learning model that includes a combination of PiQuant, metabolomics, lipidomics, and CA-19-9. The respective AUC values are provided. 【0275】 【Figure 142】 FIG. 142 shows the integrated multi-omics approach described herein. 【0276】 【Figure 143】 FIG. 143 shows several research samples and the amount of protein groups per sample. 【0277】 【Figure 144】 FIG. 144 shows several coefficient of variation (CV) values. 【0278】 【Figure 145】 FIG. 145 shows the number of protein groups detected across the subject samples. 【0279】 【Figure 146】Figure 146 shows the number of unique protein groups detected in the nanoparticle panel and in individual nanoparticles. 【0280】 【Figure 147】 Figure 147 shows the detected proteins as dots and their concentrations from the Human Plasma Proteome Project (HPPP). Proteins with top Open Targets (OT) scores are shown. 【0281】 【Figure 148A】 Figure 148A includes a ROC plot showing classifier performance for subjects with NSCLC at all stages. 【0282】 【Figure 148B】 Figure 148B includes aspects of the sensitivity of the analysis of RNA-seq data, metabolome data, protein data, and data of the combination of RNA-seq + metabolome + protein for subjects with NSCLC at all stages. 【0283】 【Figure 149A】 Figure 149A includes a ROC plot showing classifier performance for subjects with stage I NSCLC. 【0284】 【Figure 149B】 Figure 149B includes aspects of the sensitivity of the analysis of RNA-seq data, metabolome data, protein data, and data of the combination of RNA-seq + metabolome + protein for subjects with stage I NSCLC. 【0285】 【Figure 150】 Figure 150 shows the workflow for spectral library data generation. 【0286】 【Figure 151A】 Figures 151A - D show a comparison of strategies for spectral library creation. 【Figure 151B】Figures 151A to D show a comparison of strategies for creating a spectral library. 【Figure 151C】 Figures 151A to D show a comparison of strategies for creating a spectral library. 【Figure 151D】 Figures 151A to D show a comparison of strategies for creating a spectral library. 【0287】 【Figure 152】 Figure 152 shows a Jaccard index comparison for pairs of spectral libraries. 【0288】 【Figure 153A】 Figure 153A shows a combination for pairs of spectral libraries. 【0289】 【Figure 153B】 Figure 153B shows a comparison of three or more spectral libraries. 【0290】 【Figure 153C】 Figure 153C shows library construction efficiency. 【0291】 【Figure 154】 Figure 154 shows the application of the largest spectral library to 40 clinical samples. 【0292】 【Figure 155】 Figure 155 shows the effect of the size of the spectral library on Zeno-SWATH data. 【0293】 【Figure 156】 Figure 156 shows an experimental workflow using timTOF from sample processing to data collection and finally data analysis. 【0294】 【Figure 157】 Figure 157 shows the qualitative performance of timsTOF HT versus timsTOF Pro2 across a wide range of plasma peptide loading masses and LC gradients. 【0295】 Figure 158 shows the comparison between three consecutive measurements of precursors measured with different LC gradients between timsTOF HT and timsTOF Pro2. 【0296】 【Figure 158A】 Figure 158A shows the comparison between three consecutive measurements of precursors from undiluted plasma measured with different LC gradients between timsTOF HT and timsTOF Pro2. 【0297】 【Figure 158B】 Figure 158B shows the comparison between three consecutive measurements of precursors from proteograph-treated plasma (NP2) measured with different LC gradients between timsTOF HT and timsTOF Pro2. 【0298】 Figure 159 shows the comparison between the quantitative linear ranges of timsTOF HT and timsTOF Pro2. 【0299】 【Figure 159A】 Figure 159A shows a representative total ion chromatogram (TIC) of a single PG-NP2 replicate loading of 100 - 1200 ng with a gradient of 60 SPD between timsTOF HT and timsTOF Pro2. 【0300】 【Figure 159B】 Figure 159B shows the distribution of the ratio of precursor MS2 peak areas (average of three replicates) quantified by timsTOF HT and timsTOF Pro2. 【0301】 【Figure 159C】 Figure 159C shows the R-squared distribution of precursors quantified in three consecutive measurements of each peptide loading mass within the PG NP2 peptide loading ranges of 100 - 600 ng, 900 ng, or 1200 ng with a gradient of 60 SPD in timsTOF HT and timsTOF Pro2. 【0302】 【Figure 160】 Figure 160 shows the significance of cancer biomarkers detected in plasma of control and case samples between timsTOF HT and timsTOF Pro2. 【0303】 【Figure 161】 Figure 161 shows the overall experimental workflow using ZenoTOF from sample preparation to sample processing and further to data analysis. 【0304】 【Figure 162】 Figure 162 shows the workflow used for the harmonization of instruments for the discovery of over 3,000 target biomarkers. 【0305】 【Figure 163】 Figure 163 shows the CV% distribution of the number of precursors (number of unprocessed MS2) in within-batch reproducibility over the first 1,596 samples. 【0306】 【Figure 164A】 Figure 164A shows the CV% distribution of the number of precursors (number of unprocessed MS2) in between-batch reproducibility over the first 1,596 samples for each of the four instruments. 【0307】 【Figure 164B】 Figure 164B shows the CV% distribution of the number of precursors (number of unprocessed MS2) in between-batch reproducibility over the first 1,596 samples across the four instruments. 【0308】 【Figure 165】 Figure 165 shows the number of unique peptides as a function of detection frequency over 1,424 subjects. 【0309】 【Figure 166】 Figure 166 shows the number of protein groups as a function of detection frequency over 1,424 subjects. 【Mode for Carrying Out the Invention】 【0310】 The present disclosure provides a non-invasive method for diagnosing or excluding the presence of a disease in a subject, or the risk of developing a disease in the subject. The disease may include cancers such as pancreatic cancer, breast cancer, liver cancer, ovarian cancer, or colorectal cancer. Identifying an early disease in a subject and providing early treatment can save the subject from further disease progression. The non-invasive test can also be used to exclude the presence of a disease, thereby eliminating the need for the subject to undergo invasive tests such as biopsies that may be associated with pain and stress or pose a risk of damaging the subject. 【0311】 The present disclosure also provides a non-invasive method for detecting the presence of cancer such as pancreatic cancer in a subject or the risk of developing cancer. Identifying cancer early in a subject and providing early treatment can save the subject from further cancer development. The non-invasive test can also be used to exclude the presence of cancer, thereby eliminating the need for the subject to undergo invasive tests such as biopsies that may be associated with pain and stress or pose a risk of damaging the subject. 【0312】 The multi-omics approach can unlock the ability to detect diseases early in their development and improve the accuracy of disease detection. FIG. 1A shows some aspects of the multi-omics approach to early disease detection that can combine genomic DNA or DNA methylation information (an example of what can generally be a static indicator of risk) with molecular phenotype information obtained from proteomics or metabolomics, which can be a more dynamic indicator of function. FIG. 24 also shows some aspects that can be included in the multi-omics method and includes some examples of disease states that can be detected or evaluated. FIG. 1B shows an example of the integration of multiple omics data types. Any aspect of these figures can be used in the methods described herein. 【0313】 Figure 2A shows a non-limiting example of a method for predicting whether a subject has a disease such as cancer or is at risk of developing the disease. The analysis may include obtaining a biological fluid sample from the subject (200). The sample can be assayed or analyzed. The biological fluid sample can be any one or any combination of the biological fluids described herein. The sample can be any of the following: analyzed directly to generate data such as proteomics data (202); or contacted with the particles described herein prior to the analysis of 202 to obtain adsorbed biomolecules (203). After obtaining data from the analysis of 202, additional analysis (203) can be performed on the sample obtained from 200 or 201 to obtain additional data sets such as transcriptomics data, genomics data, metabolomics data, or combinations thereof. A classifier can be generated using the data or data sets obtained from the analysis of 202 or 203 (205). The classifier can be applied to identify the likelihood that the subject has the disease or is at risk of having the disease. The generation or application of the classifier can be further repeated or refined to improve the analysis. Figure 2B further shows some details that can be used in the methods described herein. Any of the aspects of Figure 2A or Figure 2B can be used in the methods described herein, such as classification methods. 【0314】 Furthermore, the analysis shown in FIGS. 2A or 2B can be applied before or during the treatment at any step included in FIG. 3A. For example, the evaluation or analysis can be completed early in the course of an affected patient, before, immediately after, or as part of an invasive biopsy. It is useful to screen high-risk patients before performing invasive procedures such as biopsies or invasive treatments. Generally, the opportunity where the methods described herein may be useful can be when screening high-risk patients for early detection of a disease. The methods described herein can be used for such detection with higher accuracy and convenience than other methods. In FIG. 3A, non-invasive biopsies may include medical imaging, and invasive biopsies may include obtaining a biopsy. A biopsy may be suspected of a tumor. Similar patient courses for pancreatic cancer, liver cancer, and colorectal cancer are shown in FIGS. 3B, 3C, and 41. The evaluation or analysis can be completed at any point in FIG. 3B, FIG. 3C, or FIG. 41 or before that. 【0315】 In some embodiments, the cancer detected by the methods described herein can be pancreatic cancer. The pancreatic cancer can be early-stage pancreatic cancer. In other embodiments, the pancreatic cancer can be late-stage pancreatic cancer. Samples obtained non-invasively can be used for cancer diagnosis by generating data and identifying patterns within the data related to cancers such as pancreatic cancer. Cancer diagnosis can be improved by obtaining proteomic data. Cancer diagnosis can be improved by combining and analyzing multiple types of data (e.g., multiple datasets). For example, combining multiple data types including proteomics, transcriptomics, genomics, metabolomics, or combinations thereof can improve the accuracy of predicting whether a subject has cancer. In some embodiments, the methods described herein include generating or obtaining data and using the data to predict whether a subject has cancer. Various methods of combining or analyzing data are described and the use of data for cancer evaluation is further detailed. 【0316】 In certain embodiments, a method of detecting cancer may include additional screening or diagnostic methods, such as a computed tomography (CT) scan indicative of pancreatic cancer, a magnetic resonance imaging (MRI) scan indicative of pancreatic cancer, a positron emission tomography (PET) scan indicative of pancreatic cancer, an ultrasound examination indicative of pancreatic cancer, an endoscopic retrograde cholangiopancreatography indicative of pancreatic cancer, an angiography indicative of pancreatic cancer, a liver function test (LFT) indicative of pancreatic cancer, an increase in carcinoembryonic antigen (CEA) levels relative to a control or baseline measurement, an increase in carbohydrate antigen (CA) 19-9 levels relative to a control or baseline measurement, or combinations thereof. In some embodiments, a method of detecting pancreatic cancer may include identifying symptoms of the subject such as jaundice, abdominal pain, gallbladder or liver enlargement, thrombosis, digestive disorders, or depression, or combinations thereof. 【0317】 When the disease is pancreatic cancer, there is an opportunity to screen high-risk patients prior to biopsy or pancreatoscopy. For example, the main opportunities for using the methods described herein include screening high-risk pancreatic cancer patients for early detection with improved accuracy and convenience. In the course of a liver cancer patient, there is an opportunity to screen high-risk liver cancer patients prior to biopsy. For example, the main opportunities for using the methods described herein may include improving the determination regarding indeterminate liver nodules to determine whether a biopsy is necessary. Another opportunity may include surveillance or diagnosis of small, low-risk nodules, or follow-up (e.g., 3 to 6 months) to track the progression of small nodules. In the course of a colorectal cancer (CRC) patient, there may be an opportunity to screen high-risk patients prior to a colonoscopy. There may be another opportunity to improve the decisions made regarding imaging or biopsy procedures. 【0318】 Non-invasively obtained samples can be used for disease diagnosis by generating omics data and identifying patterns within the omics data related to the disease. Disease diagnosis can be improved by analyzing a combination of multiple types of data (e.g., multiple datasets such as omics datasets). For example, combining multiple data types can improve the accuracy of predicting whether a subject has a particular disease. When individual datasets independently indicate errors or are not completely overlapping, the combined data may be more accurate than the individual datasets. The methods described herein include generating or obtaining multi-omics data and using the multi-omics data to make predictions about whether a subject has a disease. Various methods for combining or analyzing multi-omics data will be described. The use of multi-omics data and the evaluation of diseases will be described in more detail. 【0319】 Several methods can be used to classify pulmonary nodules. Pulmonary nodules can be either benign or malignant. Malignant pulmonary nodules can progress rapidly and may progress to lung cancer, a common and fatal cancer. There is a need to improve the identification of malignant and benign pulmonary nodules. On the one hand, early diagnosis of malignant pulmonary nodules can lead to early treatment regimens for subjects with malignant pulmonary nodules and can improve their prognosis. On the other hand, non-invasive diagnosis of benign or non-malignant pulmonary nodules helps avoid obtaining a lung biopsy, which can be costly and invasive, and thus may be more favorable for subjects with non-malignant pulmonary nodules. 【0320】 However, the development of useful clinical tests for diagnosing and analyzing whether a pulmonary nodule is benign or malignant has hardly progressed. Imaging methods often result in a high rate of misdiagnosis (such as false positives). Usually, smaller nodules are not detected by these imaging methods. Other non-invasive methods, such as biomarker screening, also have limitations. Considering that plasma contacts many tissues in the body, the proteins in plasma can be a useful biomarker discovery matrix. However, several factors, including a wide range of concentrations (e.g., on the order of 10 digits), can cause problems with plasma proteins. Complex biochemical workflows have tried to avoid these issues, but may not be practical for discovery studies of a size sufficient to ensure validation and reproducibility. Alternatively, biomarker research has been limited to the evaluation or re-evaluation of existing markers without substantial improvement in clinical performance. Therefore, there remains a need for a method for diagnosing or screening for the presence of benign or malignant pulmonary nodules based on the analysis of biomarkers in a biological fluid sample. The methods described herein can address this need. 【0321】 Disclosed herein are methods including obtaining biomolecular data. The biomolecular data may include multi-omics data. The method can include generating or receiving the data and then performing an evaluation using a classifier. The evaluation may include applying the classifier, identifying a disease, ruling out the presence of a disease, predicting the likelihood of a disease, or selecting a treatment for a disease. 【0322】 Disclosed herein are methods including using multi-omics data to evaluate a biological state. Disclosed herein are methods including using a combination of protein markers, genetics, and metabolic markers to evaluate a biological state. The biological state can include a disease such as cancer. The biological state can include a healthy state. The biological state can include a disease-free state. 【0323】 Disclosed herein are methods that include obtaining a multi-omics database that includes multi-omics data generated from a biological fluid sample. The sample can be from a population having various disease states and patient characteristics. Some aspects include querying the multi-omics database. Querying can be identifying a biomarker or set of biomarkers that can distinguish individuals in a population having a first disease state or patient characteristic from other individuals in a population having a second disease state or patient characteristic. The multi-omics data can include a combination of proteomics, metabolomics, lipidomics, transcriptomics, fragmentomics, methylomics, or genomics. 【0324】 Disclosed herein are methods that include obtaining multi-omics data from one or more biological fluid samples of a subject identified as having a lung nodule and applying a classifier to the multi-omics data to evaluate the lung nodule. The evaluation can be for determining whether the lung nodule is cancerous or non-cancerous. The evaluation can be for ruling out lung cancer. 【0325】 Disclosed herein are methods that include obtaining multi-omics data from one or more biological fluid samples of a subject suspected of having pancreatic cancer and applying a classifier to the multi-omics data to evaluate the subject. The evaluation can include determining or indicating the likelihood that the subject has pancreatic cancer. 【0326】 Some aspects relate to sample preparation. Some aspects include preparing a sample for a method disclosed herein. Some methods include preparing multiple samples. 【0327】 disease The methods described herein can be used to assess a disease state. The methods described herein can be used to predict or identify a disease state. The disease state may include a disease or disorder such as cancer. Examples of cancer include lung cancer, colorectal cancer, pancreatic cancer, liver cancer, ovarian cancer, breast cancer, prostate cancer, melanoma, bladder cancer, lymphoma, leukemia, kidney cancer, or uterine cancer. In some embodiments, the cancer is breast cancer. The disease may include a disorder. The disease state may include having a co-morbid disease associated with the disease or disorder. References to whether a subject has a disease state may include the subject being healthy. In a healthy state, the disease state can be excluded. For example, in a healthy state, having cancer can be excluded. In a disease state, being healthy can be excluded. 【0328】 These methods can be useful for cancer diagnosis. These methods can be useful for cancer screening. This method can be useful for cancer treatment. This method can include assaying proteins in a biological fluid sample obtained from or suspected of having a nodule, such as a lung nodule, to obtain protein measurements. This method can include applying a classifier to the protein measurements and thereby identifying the protein measurements as indicative of whether the lung nodule is cancerous or non-cancerous. Optionally, the classifier is generated using proteomics data obtained by contacting a training sample with particles such that the particles adsorb the proteins in the training sample and assaying the proteins adsorbed to the particles. Some embodiments include obtaining or receiving a biological fluid sample from a subject. 【0329】 In some aspects, the cancer detected by the methods described herein can be pancreatic cancer, liver cancer, ovarian cancer, or colorectal cancer. The diagnosis of cancer may be improved by obtaining proteomic data or other omics data (such as lipidomics data). The diagnosis of cancer can be improved by analyzing a combination of multiple types of data (e.g., multiple datasets). For example, combining multiple data types including proteomics, transcriptomics, genomics, metabolomics, or combinations thereof can improve the accuracy of predicting whether a subject has cancer. In some aspects, the methods described herein include generating or obtaining data and using the data to predict whether a subject has cancer. This method may include identifying the type of cancer (e.g., liver cancer versus ovarian cancer). Various methods of combining or analyzing data are described, and the use of data for cancer assessment is further detailed. 【0330】 The cancer can be early or late stage. Examples of early stage cancer include stage I. Early stage may include stage I or II. Early stage may include stage I, II, or III. Examples of late stage cancer may include stage 4. 【0331】 The cancer may include pancreatic cancer. The pancreatic cancer can be early-stage pancreatic cancer. In other embodiments, the pancreatic cancer can be late-stage pancreatic cancer. Samples obtained non-invasively can be used for cancer diagnosis by generating data and identifying patterns within the data related to cancers such as pancreatic cancer. In certain embodiments, the method of detecting cancer may include additional screening or diagnostic methods, by way of example, a computed tomography (CT) scan indicative of pancreatic cancer, a magnetic resonance imaging (MRI) scan indicative of pancreatic cancer, a positron emission tomography (PET) scan indicative of pancreatic cancer, an ultrasound examination indicative of pancreatic cancer, an endoscopic retrograde cholangiopancreatography indicative of pancreatic cancer, an angiography indicative of pancreatic cancer, a liver function test (LFT) indicative of pancreatic cancer, an increase in carcinoembryonic antigen (CEA) levels relative to control or baseline measurements, an increase in carbohydrate antigen (CA) 19-9 levels relative to control or baseline measurements, or combinations thereof. In some embodiments, the method of detecting pancreatic cancer may include identifying symptoms of the subject such as jaundice, abdominal pain, gallbladder or liver enlargement, thrombosis, digestive disorders, or depression, or combinations thereof. Any of these embodiments can be used in identifying a subject at risk of having pancreatic cancer. 【0332】 The cancer may include liver cancer. In some embodiments, the cancer detected by the methods described herein can be liver cancer. The liver cancer can be early-stage liver cancer. In other embodiments, the liver cancer can be late-stage liver cancer. In some cases, the liver cancer can be stage I, II, III, or IV liver cancer. In some cases, the stage of the liver cancer is unknown. Non-invasively obtained samples can be used for cancer diagnosis by generating data and identifying patterns in the data related to cancers such as liver cancer. In certain embodiments, the method of detecting cancer may include additional screening or diagnostic methods, such as undergoing a dynamic contrast computed tomography (CT) scan indicative of liver cancer, a magnetic resonance imaging (MRI) scan indicative of liver cancer, a liver function test (LFT) indicative of liver cancer, an increase in bilirubin level relative to a control or baseline measurement, an increase in aminotransferase level relative to a control or baseline measurement, an increase in alkaline phosphatase level relative to a control or baseline measurement, having hypoalbuminemia, an increase in prothrombin time relative to a control or baseline measurement, an increase in alpha-fetoprotein level relative to a control or baseline measurement, or having liver nodules, or combinations thereof. In some embodiments, the method of detecting cancer may include identifying symptoms of the subject such as abdominal discomfort, pain, and tenderness, jaundice, white chalky stools, nausea, vomiting, bruising, or easy bleeding, fatigue, or malaise, or combinations thereof. Any of these embodiments can be used in identifying a subject at risk of having liver cancer. 【0333】 The cancer may include ovarian cancer. In some embodiments, the cancer detected by the methods described herein can be ovarian cancer. The ovarian cancer can be early-stage ovarian cancer. In other embodiments, the ovarian cancer can be advanced-stage ovarian cancer. In some cases, the stage of the ovarian cancer may be unknown. In some embodiments, the stage of the ovarian cancer may be stage I, II, III, or IV. Non-invasively obtained samples can be used for cancer diagnosis by generating data and identifying patterns within the data related to cancers such as ovarian cancer. In certain embodiments, the method of detecting cancer may include additional screening or diagnostic methods, such as undergoing a computed tomography (CT) scan indicative of ovarian cancer, a magnetic resonance imaging (MRI) scan indicative of ovarian cancer, a positron emission tomography (PET) scan indicative of ovarian cancer, a transvaginal ultrasound examination indicative of ovarian cancer, an increase in cancer antigen (CA)-125 levels relative to control or baseline measurements, or having an ovarian cyst, or a combination thereof. In some embodiments, the method of detecting cancer may include identifying symptoms of the subject such as pelvic pressure, lower abdominal pain, vaginal bleeding, weight gain, weight loss, abnormal menstrual cycles, unexplained back pain that worsens over time, increased urine output, gas, nausea, vomiting, or loss of appetite, or a combination thereof. Any of these embodiments can be used in identifying a subject at risk of having ovarian cancer. 【0334】 The cancer may include colorectal cancer or colorectal cancer (CRC). In some embodiments, the cancer detected by the methods described herein can be colorectal cancer. The colorectal cancer may be early-stage colorectal cancer. In other embodiments, the colorectal cancer can be advanced colorectal cancer. Samples obtained non-invasively can be used for cancer diagnosis by generating data and identifying patterns within the data related to cancer, such as colorectal cancer. The diagnosis of cancer can be improved by obtaining proteomic data. In certain embodiments, the method of detecting cancer may include additional screening or diagnostic methods, by way of example, a computed tomography (CT) scan indicating an indicator of colorectal cancer, a liver function test (LFT) indicating an indicator of colorectal cancer, measurement of carcinoembryonic antigen (CEA) levels relative to control or baseline measurements, measurement of blood in the stool, performance of a fecal immunochemical test (FIT), or a combination thereof. Any of these embodiments can be used in identifying a subject at risk of having colorectal cancer. For example, a subject identified as being at risk of having colorectal cancer can be identified as being at risk by one of these methods. By the non-invasive methods described herein, a patient without colorectal cancer can avoid further invasive tests or procedures, such as colonoscopy or cancer biopsy, or colorectal cancer treatment procedures. On the other hand, the non-invasive methods described herein can be used to identify a person likely to have colorectal cancer and confirm that the patient needs to undergo further tests (e.g., invasive tests) or treatment procedures. Colorectal cancer can be an example of colorectal cancer (CRC). References or teachings in this specification related to colorectal cancer can be applied to CRC, or vice versa. 【0335】 Cancer may include lung cancer. An example of lung cancer is non-small cell lung cancer (NSCLC). An example of lung cancer is small cell lung cancer. A method for diagnosing lung nodules is disclosed. This method can be useful for diagnosing, treating, or screening patients who have had a lung nodule identified by computed tomography (CT) scan and who have not undergone a lung biopsy. This method can be useful for informing a physician about the probability that a lung nodule is benign or malignant. With test results from such a method, a physician can avoid having a patient undergo an unnecessary biopsy. For example, this method can be used as a rule-out test. With test results from such a method, a physician can identify patients who should undergo a biopsy. For example, this method can be used as a rule-in test. 【0336】 A diagnostic method for identifying candidates for CT imaging is disclosed. This method can be useful for diagnosing, treating, or screening patients who may be candidates for CT imaging. This method can be useful for high-risk patients (e.g., as defined by the USPSTF or other organizations) who are candidates for lung cancer screening but who have not yet had a CT scan. This method can inform a physician of the probability that a patient has lung cancer. Thus, this method can inform a physician of the urgency of a patient's lungs or the need to obtain a CT scan of the patient. Such a method can be useful for high-risk patients, such as those who do not follow other CT screening methods. This method can improve patient selection or compliance for CT imaging. This method can improve patient selection or compliance for biopsy. 【0337】 A method for recurrent monitoring is disclosed. This method can be useful for monitoring patients with potentially resectable lung cancer. This method can be useful for monitoring patients who have received postoperative treatment interventions. This method can be useful for monitoring patients who are undergoing adjuvant chemotherapy or radiation therapy interventions. This method can be useful for detecting cancer recurrence prior to a CT scan or other medical imaging. This method can be useful for recurrence surveillance testing. This method can be adjusted or developed according to the patient's treatment method. 【0338】 Described herein is assaying a protein in a biological fluid sample obtained from a subject having or suspected of having a lung nodule to obtain a protein measurement; and applying a classifier to the protein measurement to thereby identify the protein measurement as indicative of the lung nodule being cancerous or non-cancerous, wherein the classifier is generated using proteomics data obtained by contacting a training sample with particles such that the particles adsorb the protein in the training sample and assaying the protein adsorbed to the particles. This method can be useful for cancer diagnosis or screening. 【0339】 Described herein is a method comprising obtaining a biological fluid sample from a subject having a lung nodule; contacting the biological fluid sample with particles such that the particles adsorb biomolecules including proteins; assaying the biomolecules adsorbed to the particles to generate proteomics data; and classifying the proteomics data as indicative of whether the lung nodule is cancerous or non-cancerous. This method can be useful for cancer diagnosis or screening. 【0340】 Described herein is a method for determining a lung nodule-related condition in a sample obtained from a subject. In some embodiments, the lung nodule-related condition includes the presence or absence of a lung nodule in the subject. In some embodiments, the lung nodule-related condition includes determining whether the lung nodule is benign or malignant. In some embodiments, the method includes screening for a lung nodule-related condition by assaying biomarkers in a sample obtained from the subject. In some embodiments, the biomarker includes at least one protein in the sample. In some embodiments, the sample is a biological fluid sample. In some embodiments, the biological fluid sample is contacted with the particles described herein to adsorb proteins in the biological fluid sample. In some embodiments, the method includes obtaining a protein measurement value of the protein in the sample. In some embodiments, the method includes applying a classifier to the protein measurement value, thereby identifying the protein measurement value as indicative of whether the lung nodule is cancerous or non-cancerous. In some embodiments, the classifier is generated using proteomics data obtained by contacting a training sample with the particles such that the particles adsorb proteins in the training sample. The adsorbed proteins can then be assayed by the methods described herein. In some embodiments, the subject is suspected of having a lung nodule or is identified as having a lung nodule by the imaging methods described herein. In some embodiments, the report is generated based on the identification of a protein measurement value indicative of whether the lung nodule is cancerous or non-cancerous. In some embodiments, the report indicates the likelihood or an indicator of whether the lung nodule is cancerous or non-cancerous. In some embodiments, the report indicates that the lung nodule is cancerous. In some embodiments, the report indicates that the lung nodule includes non-small cell lung cancer (NSCLC). In some embodiments, the method described herein generates a classifier that includes a feature indicative of a protein measurement value indicative of whether the lung nodule is cancerous or non-cancerous.In some embodiments, the feature quantity includes a control protein measurement value, a mass spectrum, an m / z ratio, a chromatography result, an immunoassay result, or a light or fluorescence intensity. In some embodiments, the classifier is trained using any one of the computational methods or machine learning methods described herein. 【0341】 As described herein, in some embodiments, a method for recommending lung cancer treatment to a subject is described when it is determined that the subject has a malignant lung nodule based on the analysis of the protein measurement values described herein. In some embodiments, the protein measurement values are classified as indicating that the lung nodule is cancerous. 【0342】 As described herein, in some aspects, methods useful for diagnosing, screening, or treating a subject are disclosed. Some aspects include assaying proteins in a biological fluid sample obtained from a subject suspected of having a lung nodule to obtain protein measurement values. Some aspects include applying a classifier to the protein measurement values. Some aspects include identifying protein measurement values as indicating that the subject has a lung nodule. In some embodiments, the classifier is generated using proteomics data obtained by contacting a training sample with particles such that the particles adsorb the proteins in the training sample and assaying the proteins adsorbed to the particles. 【0343】 In some aspects, methods useful for diagnosing, screening, or treating a subject are disclosed herein. Some aspects include assaying proteins in a biological fluid sample obtained from a subject suspected of having lung cancer to obtain protein measurements. Some aspects include applying a classifier to the protein measurements. Some aspects include identifying protein measurements as indicative of the subject having lung cancer. In some aspects, the classifier is generated using proteomics data obtained by contacting a training sample with particles such that the particles adsorb proteins in the training sample and assaying the proteins adsorbed to the particles. 【0344】 In some aspects, methods useful for diagnosing, screening, or treating a subject are disclosed herein. Some aspects include obtaining a biological fluid sample from a subject suspected of having a lung nodule. Some aspects include contacting the biological fluid sample with particles such that the particles adsorb biomolecules containing proteins. Some aspects include assaying the biomolecules adsorbed to the particles to generate proteomics data. Some aspects include classifying the proteomics data based on the proteomics data as indicative of the subject having a lung nodule or not having a lung nodule. 【0345】 In some aspects, methods useful for diagnosing, screening, or treating a subject are disclosed herein. Some aspects include obtaining a biological fluid sample from a subject suspected of having lung cancer. Some aspects include contacting the biological fluid sample with particles such that the particles adsorb biomolecules containing proteins. Some aspects include assaying the biomolecules adsorbed to the particles to generate proteomics data. Some aspects include classifying the proteomics data based on the proteomics data as indicative of the subject having lung cancer or not having lung cancer. 【0346】 In some aspects, methods useful for monitoring a subject are disclosed herein. Some aspects include obtaining a biological fluid sample from a subject at risk of lung cancer recurrence. Some aspects include contacting the biological fluid sample with particles such that the particles adsorb biomolecules including proteins to the particles. Some aspects include assaying the biomolecules adsorbed to the particles to generate proteomics data. Some aspects include classifying the proteomics data, based on the proteomics data, as indicative of the subject having lung cancer recurrence or as not indicative of the subject having lung cancer recurrence. In some aspects, the subject has received a lung cancer treatment such as chemotherapy, radiation therapy, or surgery. In some aspects, the cancer may be resectable. In some aspects, the lung cancer includes NSCLC. 【0347】 In some cases, a lung nodule is described as malignant or cancerous. The terms malignant and cancerous may be used interchangeably. A malignant or cancerous lung nodule may be referred to as lung cancer, and vice versa. In some cases, a lung nodule is described as benign or non-cancerous. The terms benign and non-cancerous may be used interchangeably. 【0348】 Samples and Subjects Some aspects relate to a subject. For example, the methods described herein can be used to evaluate a subject or a sample from a subject. Multi-omics data can be generated from a sample of the subject. 【0349】 Using the methods described herein, subjects at high risk of having or having a disease such as cancer can be identified. The subject may have lung cancer, pancreatic cancer, liver cancer, ovarian cancer, or colorectal cancer. Cancer may include adenocarcinoma, such as pancreatic adenocarcinoma. The subject may have cancer. The subject may not have cancer. The subject may have pancreatic cancer, liver cancer, ovarian cancer, or colorectal cancer. The subject may not have pancreatic cancer, liver cancer, ovarian cancer, or colorectal cancer. The subject may be at risk of having pancreatic cancer, liver cancer, ovarian cancer, or colorectal cancer. The subject may have a tumor (e.g., nodule or cyst) in the pancreas. The subject may have a tumor (e.g., nodule) in the liver. Liver cancer may include hepatocellular carcinoma (HCC). Liver cancer may include stage I, stage II, stage III, or stage IV liver cancer. The subject may have a tumor (e.g., nodule or cyst) in one or both ovaries. Ovarian cancer may include stage I, stage II, stage III, or stage IV ovarian cancer. Ovarian cancer may include stage III ovarian cancer. Ovarian cancer may include stage IV ovarian cancer. The subject may have a tumor (e.g., nodule) in the large intestine. The subject may have a lung nodule and may have cancer. The subject may be at risk of having breast cancer. The subject may have a tumor (e.g., nodule or cyst) in the breast. 【0350】 Samples can be obtained from a subject for the purpose of identifying cancer in the subject. The subject may be suspected of having cancer or may not be suspected of having cancer. This method can be used to confirm or refute the suspicion of cancer. 【0351】 The data described in this specification can be generated from a sample of interest. The sample can be a biological fluid sample or a tumor sample (e.g., a biopsied abnormal growth from a subject). Examples of biological fluids include blood, serum, or plasma. The sample may include a blood sample. The sample may include a serum sample. The sample may include a plasma sample. One or more biological fluid samples can include a blood, serum, or plasma sample. Other examples of biological fluids include urine, tears, semen, milk, vaginal fluid, mucus, saliva, sweat, or a cell homogenate. 【0352】 The sample can be obtained from a subject for the purpose of identifying a disease state in the subject. The subject may be suspected of having a disease state or may not be suspected of having a disease state. This method can be used to confirm or refute a suspicion of a disease state. In some embodiments, a sample from a subject is used to determine whether a tumor, nodule (e.g., a lung nodule), or cyst is cancerous or non-cancerous. 【0353】 A biological fluid sample can be obtained from a subject. For example, a blood, serum, or plasma sample can be obtained from a subject by venipuncture. Other methods of obtaining a biological fluid sample include aspiration or collection with a swab. 【0354】 The biological fluid sample can be cell-free or substantially cell-free. To obtain a cell-free or substantially cell-free biological fluid sample, the biological fluid can undergo sample preparation methods such as centrifugation and pellet removal. 【0355】 Non-biological fluid samples can be obtained from a subject or patient. For example, the sample may include a tissue sample. Some examples of organs or tissues that can be sampled include lung, large intestine, pancreas, liver, breast, or ovarian tissue. The sample may include a tumor taken from an organ or tissue of the subject. The tumor may be suspected of being cancerous. The tumor may include nodules (e.g., colorectal nodules or liver nodules). The tumor may include cysts (e.g., ovarian cysts). Nodules or cysts can be identified by a physician as having a high or low risk of being cancerous prior to performing the methods described herein. The tumor may be biopsied, for example, by a needle biopsy procedure. The needle biopsy procedure may include inserting a thin needle into the liver from the subject's abdomen to obtain a tissue sample, and then this tissue sample can be examined microscopically for signs of cancer. The sample may include a cell sample. The sample may include a homogenate of cells or tissues. The sample may include the supernatant of a centrifuged homogenate of cells or tissues. 【0356】 The sample may include lung tissue. The sample may include large intestine tissue. The sample may include pancreas tissue. The sample may include liver tissue. The sample may include breast tissue. The sample may include ovarian tissue. The tissue may be cancerous. The tissue may be non-cancerous. The tissue may be suspected of being cancerous. The tissue may be malignant. The tissue may be non-malignant. The tissue may be suspected of being malignant. 【0357】 The sample (e.g., a biological fluid or tissue sample) can be obtained from the subject during any stage of the screening procedure, such as before, during, or after the stage shown in Figure 3A. The sample can be obtained from the subject before or during the stage when the subject is a candidate for biopsy, pancreatoscopy, or colonoscopy for early detection of the disease. The sample can be obtained before or during non-invasive diagnostic tests, invasive diagnostic tests, treatment, or monitoring stages. 【0358】 Data can be generated from a single sample or from multiple samples. Data from multiple samples can be obtained from the same subject. In some cases, different data types are obtained from samples collected in a different manner or from samples collected in separate containers. Samples may be collected in a container containing one or more reagents such as a preservative reagent or a biomolecule isolation reagent. Some examples of reagents include heparin, ethylenediaminetetraacetic acid (EDTA), citrate, an anti-lysing agent, or a combination of reagents. Samples from a subject may be collected in multiple containers containing different reagents for purposes such as preserving or separating different types of biomolecules. Samples may be collected in a container that does not contain a reagent. Samples can be collected simultaneously (e.g., at the same time or on the same day) or at different times. Samples can be frozen, refrigerated, heated, or stored at room temperature. 【0359】 Using the methods described herein, a subject can be identified as likely or not likely to have a disease state. Examples of disease states can include cancers such as pancreatic cancer, liver cancer, ovarian cancer, or colorectal cancer. Some aspects of the present disclosure include identifying whether a pulmonary nodule of a subject is cancerous or non-cancerous. The pulmonary nodule may be present in the lungs of the subject. The subject can be identified as having a pulmonary nodule. In some aspects, the subject has multiple pulmonary nodules. The subject may have lung cancer. The subject may be at risk of lung cancer. The subject may have a pulmonary complication. The subject may have a co-existing disease described herein. The subject may experience difficulty breathing. The subject may have fluid in the lungs. 【0360】 In some cases, the subject is monitored. For example, using information about the likelihood that the subject has a disease state, it can be determined to monitor the subject without providing treatment to the subject. In other situations, the subject can be monitored to determine whether the subject's disease state improves while undergoing treatment. In some embodiments, a subject having a pulmonary nodule can be monitored to determine the progression of the pulmonary nodule. The pulmonary nodules of the subject can be monitored. The subject can be treated as described herein. 【0361】 The subject may be a vertebrate. The subject may be a mammal. Mammals may include rats, mice, gerbils, guinea pigs, or hamsters. Mammals may include foxes, bears, dogs, monkeys, cows, pigs, or sheep. The subject may be a primate. Primates may include apes or monkeys. Primates may include chimpanzees, cynomolgus monkeys, bonobos, orangutans, or baboons. The subject can be a human. The subject may be an adult (e.g., at least 18 years old). The subject may be male. The subject may be female. The subject may have a disease state. For example, the subject may have a disease or disorder, a co - morbidity of a disease or disorder, or may be healthy. 【0362】 The methods described herein may include the use of samples such as biological samples. For example, the method may include determining one or more biomarker measurements in the sample. The biological sample can be from a subject such as a subject having a pulmonary nodule. The biological sample may include a blood sample from which red blood cells have been removed. For example, the biological sample may include a plasma sample. The biological sample may include a serum sample. The biological sample may include blood or a blood component. The biological sample may include a blood sample. The samples described or used herein can be from the subjects described herein, such as a subject having an identified pulmonary nodule. 【0363】 The sample is consistent with the methods disclosed herein for assessing the presence or absence of one or more biomarkers associated with the presence or malignancy of a lung nodule. The subject may be human or a non-human animal. The biological sample may be a biological fluid. For example, the biological fluid may be plasma, serum, CSF, urine, tears, cell lysate, tissue lysate, cell homogenate, tissue homogenate, nipple aspirate, fecal sample, synovial fluid, and whole blood, or saliva. The sample may also be a non-biological sample such as water, milk, a solvent, or something homogenized to a fluid state. The biological sample can contain a plurality of proteins or proteomics data, which can be analyzed after adsorption of the proteins to the surface of various types of particles in the panel and subsequent digestion of the protein corona. The proteomics data can include nucleic acids, peptides, or proteins. Any of the samples herein can contain a number of different analytes that can be analyzed using the methods disclosed herein. An analyte can be a protein, peptide, small molecule, nucleic acid, metabolite, lipid, or any molecule that can bind to or interact with the surface of a particle type. 【0364】 The sample may be a biological fluid. The biological sample may be a biological fluid sample, for example, cerebrospinal fluid (CSF), synovial fluid (SF), urine, plasma, serum, tears, gingival crevicular fluid, semen, whole blood, milk, nipple aspirate fluid, ductal lavage fluid, vaginal fluid, nasal mucus, ear fluid, gastric juice, pancreatic juice, fibrous cord fluid, lung lavage fluid, prostatic fluid, sputum, feces, bronchial lavage fluid, fluid from swabbing, bronchial aspirate, sweat, or saliva. The biological fluid may be a fluidized solid, such as a tissue homogenate, or a fluid extracted from a biological sample. The biological sample may be, for example, a tissue sample or a fine needle aspiration (FNA) sample. The biological sample may be a cell culture sample. For example, the sample that can be used in the methods disclosed herein can contain cells growing in a cell culture or can contain cell-free material collected from a cell culture. The biological fluid may be a fluidized biological sample. For example, the biological fluid may be a fluidized cell culture extract. The sample may be extracted from a fluid sample or the sample may be extracted from a solid sample. For example, the sample can contain gaseous molecules extracted from a fluidized solid (e.g., volatile organic compounds). In some embodiments, the biological fluid includes blood, plasma, or serum. 【0365】 Methods consistent with this disclosure can include collecting (e.g., isolating, concentrating, or purifying) a species from a biological sample. The species can be a biomolecule (e.g., a protein), a biopolymeric structure (e.g., a peptide aggregate or a ribosome), a cell, or a tissue. The species can be selectively collected from the biological sample. For example, the method may include the isolation of cancer cells from a tissue (e.g., as a tissue biopsy) or from a body fluid such as whole blood, plasma, or buffy coat (e.g., as a liquid biopsy). The method may include samples that do not contain cancer cells. The species may be processed prior to analysis. For example, a protein may be reduced and digested, a nucleic acid may be separated from histones, or a cell may be lysed. 【0366】 A biological sample can be obtained from or derived from a human subject. The biological sample can be stored at various temperatures (e.g., at room temperature, under refrigeration or freezing conditions, at 25 °C, 4 °C, -18 °C, -20 °C, or -80 °C) or in various storage conditions such as different suspensions (e.g., in an EDTA collection tube, a cell-free RNA collection tube, or a cell-free DNA collection tube) prior to processing. 【0367】 In some cases, the sample may be depleted prior to biomarker analysis. The sample can be depleted using commercially available kits. For example, kits that can be used to deplete the sample can be spin column-based depletion kits, albumin depletion kits, immunodepletion kits, or abundant protein depletion kits. Non-limiting examples of kits that can be used for sample depletion include the PureProteome™ Human Albumin / Immunoglobulin Depletion Kit (EMD Millipore Sigma), the ProteoPrep® Immunodepletion Albumin & IgG Kit (Millipore Sigma), the Seppro® Protein Depletion Kit (Millipore Sigma), the Top 12 Abundant Protein Depletion Spin Columns (Pierce), or the Proteome Purify™ Immunodepletion Kit (R&D Systems). Depletion can remove high-concentration biomolecules from the sample. For example, the method may include removing albumin from a plasma sample prior to low-concentration biomarker analysis. The sample may include depleted plasma. 【0368】 Generation and Use of Data The methods disclosed herein may include obtaining data such as multi-omics data generated from one or more biological fluid samples collected from a subject. The data may include biomolecular measurements such as protein measurements, transcript measurements, genetic material measurements, or metabolite measurements. The omics data may include any of the following: proteomics data, genomics data, transcriptomics data, or metabolomics data. This section includes several methods for generating each of these types of omics data. The methods for generating or analyzing omics data can also be applied to methods for generating or analyzing individual biomolecules or sets of biomolecules. Other types of omics data can also be generated. The description of the generation or analysis of omics data can be applied to methods for generating or analyzing individual biomolecules or sets of biomolecules that do not necessarily include omics data. The aspects described in relation to biomolecular data may relate to the measurement of biomolecules or vice versa. The data can be labeled or identified as indicative of a disease or not indicative of a disease. The data can be labeled or identified as indicative of pancreatic cancer, liver cancer, ovarian cancer, or colorectal cancer or not indicative of pancreatic cancer, liver cancer, ovarian cancer, or colorectal cancer. The methods described herein may include obtaining multi-omics measurements, such as by performing an assay. 【0369】 The methods described herein may include generating or using omics data. Omics data may include data on all biomolecules of a particular type, such as proteins, transcripts, genetic material, or metabolites. Omics data may include data on a subset of biomolecules. For example, omics data can include data on 500 or more, 750 or more, 1000 or more, 2500 or more, 5000 or more, 10,000 or more, 25,000 or more biomolecules of a particular type. The methods described herein can include obtaining measurements of 10 or more, 20 or more, 30 or more, 40 or more, 50 or more, 75 or more, 100 or more, 250 or more, 500 or more, 750 or more, 1000 or more, 1250 or more, 2500 or more, 5000 or more, 7500 or more, 10,000 or more, 12,500 or more, 15,000 or more, 17,500 or more, 20,000 or more, 22,500 or more, or 25,000 or more biomolecules of a particular type. Any of the aforementioned numbers of biomolecules can be measured for each of multiple data types. Multi-omics includes at least 100 measurements of each of at least two types of omics data. Multi-omics includes at least 500 measurements of each of at least two types of omics data. Multi-omics includes at least 1000 measurements of each of at least two types of omics data. The data may be related to the presence, absence, or amount of a given biomolecule. Examples of data types can include data on lipids, proteins, peptides, transcripts, mRNA, miRNA, DNA sequences, methylation, or metabolites. 【0370】 Deep proteome coverage is advantageous for multi-omics approaches. New technologies and sample availability address past challenges in scaling proteomics. Some of these challenges include access to large-scale sample cohorts that are well-collected and annotated for specific clinical problems, technical challenges associated with plasma proteomics that can limit translation to the clinic, such as reproducibility, throughput, and depth of coverage, and the reproducible measurement and integration of multi-omics datasets that provide new insights into cancer biology. 【0371】 The concepts described herein may help address some of these challenges. For example, these concerns can be addressed by using particles or including additional omics types. 【0372】 Methods for multi-omics analysis are disclosed herein. "Multi-omic(s)" or "multiomic(s)" may include analytical approaches for large-scale analysis of biomolecules, and the datasets are multiple omes such as the proteome, genome, transcriptome, lipidome, and metabolome. Non-limiting examples of multi-omics data can include proteomics data, genomics data, lipidomics data, glycomics data, transcriptomics data, or metabolomics data. "Biomolecule" in the "biomolecule corona" can refer to any molecule or biological component that can be produced by or present in a biological tissue. Non-limiting examples of biomolecules include proteins (protein corona), polypeptides, polysaccharides, sugars, lipids, lipoproteins, metabolites, oligonucleotides, nucleic acids (DNA, RNA, microRNA, plasmid, single-stranded nucleic acid, double-stranded nucleic acid), metabolome, and small molecules, such as primary metabolites, secondary metabolites, and other natural products, or any combination thereof. In some embodiments, the biomolecule is selected from the group consisting of proteins, nucleic acids, lipids, and metabolites. 【0373】 Some of the aspects that may be included in a multi-omics strategy may include a well-defined disease biobank with multiple sample types optimized for multi-omics measurements, the development and optimization of new proteomics technologies to improve proteome coverage and throughput without sacrificing reproducibility, or a non-biased multi-omics platform that incorporates state-of-the-art instrumentation and advanced machine learning analysis to transform the early detection of complex diseases. 【0374】 Proteomics data Data such as the multi-omics data described herein may include protein data or proteomics data. Proteomics data may include data related to proteins, peptides, or proteoforms. This data may include peptides only, proteins only, or a combination of both. An example of a peptide is an amino acid chain. An example of a protein is a peptide or a combination of peptides. For example, a protein can include one, two, or more peptides bound together. The protein may be a secreted protein. Proteomics data may include data related to various proteoforms. Proteoforms can include various forms of proteins generated from the genome with various sequence variations, splice isoforms, or post-translational modifications. Proteomics data may be generated using a non-biased, non-targeted approach or may include a specific set of proteins. Aspects described in relation to proteomics data may be the case in relation to protein data and vice versa. 【0375】 Proteomics data may include information regarding the presence, absence, or amount of various proteins and peptides. For example, proteomics data may include the amount of a protein. The amount of a protein can be expressed as the concentration or amount of the protein, such as the concentration of a protein in a biological fluid. The amount of a protein may be compared to another protein or another biomolecule. Proteomics data may include information regarding the presence of a protein or peptide. Proteomics data may include information regarding the absence of a protein or peptide. Proteomics data can be distinguished by subtype, and each subtype includes different types of proteins, peptides, or proteoforms. 【0376】 Proteomics data typically includes data regarding a large number of proteins or peptides. For example, proteomics data may include information regarding the presence, absence, or amount of 1000 or more proteins or peptides. In some cases, proteomics data may include information regarding the presence, absence, or amount of 5000, 10,000, 20,000, or more peptides, proteins, or proteoforms. Proteomics data can include up to approximately 1 million proteoforms. Proteomics data may include a range of proteins, peptides, or proteoforms defined by any of the aforementioned numbers of proteins, peptides, or proteoforms. Some examples of proteins or peptides that can be included in proteomics data are shown in FIGS. 6, 7, 10B, or 15. 【0377】 Proteomics data can include protein information such as protein measurements in biological fluids. Some examples of protein biomarkers that can be useful in the methods disclosed herein, such as evaluating cancers like pancreatic cancer, are included in FIG. 140D. Protein measurements can be obtained by use of internal standards. Any combination or number of such biomarkers can be included. In some cases, a biomarker is useful when its feature importance score exceeds 0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, 0.10, 0.12, or 0.14. Features are the following proteins: coagulation factor XIII A chain (F13A_HUMAN, UniProt ID P00488), aminopeptidase N (AMPN_HUMAN, UniProt ID P15144), polymeric immunoglobulin receptor (PIGR_HUMAN, UniProt ID P01833), anthrax toxin receptor 2 (ANTR2_HUMAN, UniProt ID P58335), protein S100-A8 (S10A8_HUMAN, UniProt ID P05109), leucine-rich alpha-2-glycoprotein (A2GL_HUMAN, UniProt ID P02750), apolipoprotein M (APOM_HUMAN, UniProt ID O95445), apolipoprotein C-I (APOC1_HUMAN, UniProt ID P02654), protein S100-A9 (S10A9_HUMAN, UniProt ID P06702), neuropilin-1 (NRP1_HUMAN, UniProt ID O14786), low-affinity immunoglobulin gamma Fc region receptor III-A (FCG3A_HUMAN, UniProt ID P08637), transthyretin (TTHY_HUMAN, UniProt ID P02766), cartilage acidic protein 1 (CRAC1_HUMAN, UniProt ID Q9NQ79), intercellular adhesion molecule 1 (ICAM1_HUMAN, UniProt ID P05362), CD166 antigen (CD166_HUMAN, UniProt ID Q13740), tenascin (TENA_HUMAN, UniProt IDIt may contain any one of gelsolin (GELS_HUMAN, UniProt ID P06396), tenascin (TETN_HUMAN, UniProt ID P05452), insulin-like growth factor-binding protein 2 (IBP2_HUMAN, UniProt ID P18065), intelectin-1 (ITLN1_HUMAN, UniProt ID Q8WWA0), inter-alpha-trypsin inhibitor heavy chain H3 (ITIH3_HUMAN, UniProt ID Q06033), vascular cell adhesion protein 1 (VCAM1_HUMAN, UniProt ID P19320), or apolipoprotein C-III (APOC3_HUMAN, UniProt ID P02656). The biomarker may include coagulation factor XIII A chain (F13A_HUMAN, UniProt ID P00488). The biomarker may include aminopeptidase N (AMPN_HUMAN, UniProt ID P15144). The biomarker may include polymeric immunoglobulin receptor (PIGR_HUMAN, UniProt ID P01833). The biomarker may include anthrax toxin receptor 2 (ANTR2_HUMAN, UniProt ID P58335). The biomarker may include protein S100-A8 (S10A8_HUMAN, UniProt ID P05109). The biomarker may include leucine-rich alpha-2-glycoprotein (A2GL_HUMAN, UniProt ID P02750). The biomarker may include apolipoprotein M (APOM_HUMAN, UniProt ID O95445). The biomarker may include apolipoprotein C-I (APOC1_HUMAN, UniProt ID P02654). The biomarker may include protein S100-A9 (S10A9_HUMAN, UniProt ID P06702). The biomarker may include neuropilin-1 (NRP1_HUMAN, UniProt ID O14786). The biomarker may include low-affinity immunoglobulin gamma Fc region receptor III-A (FCG3A_HUMAN, UniProt IDIt may include P08637). The biomarker may include transthyretin (TTHY_HUMAN, UniProt ID P02766). The biomarker may include cartilage acidic protein 1 (CRAC1_HUMAN, UniProt ID Q9NQ79). The biomarker may include intercellular adhesion molecule 1 (ICAM1_HUMAN, UniProt ID P05362). The biomarker may include CD166 antigen (CD166_HUMAN, UniProt ID Q13740). The biomarker may include tenascin (TENA_HUMAN, UniProt ID P24821). The biomarker may include gelsolin (GELS_HUMAN, UniProt ID P06396). The biomarker may include tenectin (TETN_HUMAN, UniProt ID P05452). The biomarker may include insulin-like growth factor binding protein 2 (IBP2_HUMAN, UniProt ID P18065). The biomarker may include intelectin-1 (ITLN1_HUMAN, UniProt ID Q8WWA0). The biomarker may include inter-alpha-trypsin inhibitor heavy chain H3 (ITIH3_HUMAN, UniProt ID Q06033). The biomarker may include vascular cell adhesion protein 1 (VCAM1_HUMAN, UniProt ID P19320). The biomarker may include complement component C9 (CO9_HUMAN, UniProt ID P02748). The biomarker may include apolipoprotein C-III (APOC3_HUMAN, UniProt ID P02656). 【0378】 Proteomics data can include protein information such as protein measurements in biological fluids. Some examples of protein biomarkers that may be useful in the methods disclosed herein, such as evaluating cancers such as pancreatic cancer, are included in FIG. 140F. Protein measurements can be obtained by use of particles, such as those described herein. Any combination or number of such biomarkers can be included. In some cases, a biomarker is useful when its feature importance score exceeds 0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07.The feature amount may include any one of the following proteins: interferon-induced transmembrane protein 3 (IFM3_HUMAN, UniProt ID Q01628), aminopeptidase N (AMPN_HUMAN, UniProt ID P15144), leucine-rich alpha-2-glycoprotein (A2GL_HUMAN, UniProt ID P02750), alpha-1-antichymotrypsin (AACT_HUMAN, UniProt ID P01011), SPARC-related modular calcium-binding protein 1 (SMOC1_HUMAN, UniProt ID Q9H4F8-2), alpha-1-antitrypsin (A1AT_HUMAN, UniProt ID P01009), pentraxin-related protein PTX3 (PTX3_HUMAN, UniProt ID P26022), cadherin-related family member 2 (CDHR2_HUMAN, UniProt ID Q9BYE9), histone H2A 2-C type (H2A2C_HUMAN, UniProt ID Q16777), anthrax toxin receptor 2 (ANTR2_HUMAN, UniProt ID P58335-4), matrilysin (MMP7_HUMAN, UniProt ID P09237), complement component C7 (CO7_HUMAN, UniProt ID P10643), annexin A2 (ANXA2_HUMAN, UniProt ID P07355-2), fibrinogen-like protein 1 (FGL1_HUMAN, UniProt ID Q08830), histone H4 (H4_HUMAN, UniProt ID P62805), or very long-chain specific acyl-CoA dehydrogenase, mitochondrial type (ACADV_HUMAN, UniProt ID P49748-3). The biomarker may include interferon-induced transmembrane protein 3 (IFM3_HUMAN, UniProt ID Q01628). The biomarker may include aminopeptidase N (AMPN_HUMAN, UniProt ID P15144). The biomarker may include leucine-rich alpha-2-glycoprotein (A2GL_HUMAN, UniProt ID P02750).The biomarker may include alpha-1-antichymotrypsin (AACT_HUMAN, UniProt ID P01011). The biomarker may include SPARC-related modular calcium-binding protein 1 (SMOC1_HUMAN, UniProt ID Q9H4F8) (e.g., Q9H4F8-2). The biomarker may include alpha-1-antitrypsin (A1AT_HUMAN, UniProt ID P01009). The biomarker may include pentraxin-related protein PTX3 (PTX3_HUMAN, UniProt ID P26022). The biomarker may include cadherin-related family member 2 (CDHR2_HUMAN, UniProt ID Q9BYE9). The biomarker may include histone H2A 2-C type (H2A2C_HUMAN, UniProt ID Q16777). The biomarker may include anthrax toxin receptor 2 (ANTR2_HUMAN, UniProt ID P58335) (e.g., P58335-4). The biomarker may include matrilysin (MMP7_HUMAN, UniProt ID P09237). The biomarker may include complement component C7 (CO7_HUMAN, UniProt ID P10643). The biomarker may include annexin A2 (ANXA2_HUMAN, UniProt ID P07355) (e.g., P07355-2). The biomarker may include fibrinogen-like protein 1 (FGL1_HUMAN, UniProt ID Q08830). The biomarker may include histone H4 (H4_HUMAN, UniProt ID P62805). The biomarker may include very long-chain specific acyl-CoA dehydrogenase, mitochondrial type (ACADV_HUMAN, UniProt ID P49748) (e.g., P49748-3). 【0379】 Proteomics data can include protein information such as protein measurements in biological fluids. Some examples of protein biomarkers that may be useful in the methods disclosed herein. Protein measurements can be obtained by the use of particles, such as those described herein. Any combination or number of such biomarkers can be included. The feature quantity can include any one of the following proteins: complement component C9 (CO9_HUMAN, UniProt ID P02748), complement C2 (CO2_HUMAN, UniProt ID P06681), CSC1-like protein 1 (CSCL1_HUMAN, UniProt ID O94886), cathepsin F (H0YD65_HUMAN, UniProt ID H0YD65), cartilage intermediate layer protein 2 (K7EPJ4_HUMAN, UniProt ID K7EPJ4), cathepsin B (E9PHZ5_HUMAN, UniProt ID E9PHZ5), progranulin (GRN_HUMAN, UniProt ID P28799), inter-alpha-trypsin inhibitor heavy chain H4 (ITIH4_HUMAN, UniProt ID Q14624), phospholipid transfer protein (PLTP_HUMAN, UniProt ID P55058), protease inhibitor 16 (PI16_HUMAN, UniProt ID Q6UXB8), or plasma serine protease inhibitor (IPSP_HUMAN, UniProt ID P05154). The biomarker may include complement C2 (CO2_HUMAN, UniProt ID P06681). The biomarker may include CSC1-like protein 1 (CSCL1_HUMAN, UniProt ID O94886). The biomarker may include cathepsin F (H0YD65_HUMAN, UniProt ID H0YD65). The biomarker may include cartilage intermediate layer protein 2 (K7EPJ4_HUMAN, UniProt ID K7EPJ4). The biomarker may include cathepsin B (E9PHZ5_HUMAN, UniProt ID E9PHZ5). The biomarker may include progranulin (GRN_HUMAN, UniProt ID P28799).The biomarker may include inter-alpha-trypsin inhibitor heavy chain H4 (ITIH4_HUMAN, UniProt ID Q14624). The biomarker may include phospholipid transfer protein (PLTP_HUMAN, UniProt ID P55058). 【0380】 Some examples of proteins that can be used as biomarkers are shown in Table 15E. One, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, or twenty of these proteins may be useful as biomarkers, for example, in the evaluation of pulmonary nodules.Any of the following proteins may be useful as such (as indicated by their UniProt ID numbers: coagulation factor XIII A-chain (F13A_HUMAN, UniProt ID P00488), endothelial protein C receptor (EPCR_HUMAN, UniProt ID Q9UNN8), insulin-like growth factor binding protein 2 (IBP2_HUMAN, UniProt ID P18065), phosphatidylcholine-sterol acyltransferase (LCAT_HUMAN, UniProt ID P04180), polymeric immunoglobulin receptor (PIGR_HUMAN, UniProt ID P01833), tenascin-X (TENX_HUMAN, UniProt ID P22105), attractin (ATRN_HUMAN, UniProt ID O75882), intelectin-1 (ITLN1_HUMAN, UniProt ID Q8WWA0), integrin beta-1 (ITB1_HUMAN, UniProt ID P05556), immunoglobulin heavy chain constant gamma 2 (IGHG2_HUMAN, UniProt ID P01859), alpha-N-acetylglucosaminidase (ANAG_HUMAN, UniProt ID P54802), hepatocyte growth factor activator (HGFA_HUMAN, UniProt ID Q04756), beta-Ala-His dipeptidase (CNDP1_HUMAN, UniProt ID Q96KN2), lumican (LUM_HUMAN, UniProt ID P51884), neurogenic locus notch homolog protein 2 (NOTC2_HUMAN, UniProt ID Q04721), synaptophysin-like protein 1 (SYPL1_HUMAN, UniProt ID Q16563), complement factor H-related protein 1 (FHR1_HUMAN, UniProt ID Q03591), coagulation factor VII (FA7_HUMAN, UniProt ID P08709), extracellular matrix protein 1 (ECM1_HUMAN, UniProt ID Q16610), or GDH / 6PGL endoplasmic reticulum bifunctional protein (G6PE_HUMAN, UniProt ID O95479).The biomarker may include coagulation factor XIII A chain (F13A_HUMAN, UniProt ID P00488). The biomarker may include endothelial protein C receptor (EPCR_HUMAN, UniProt ID Q9UNN8). The biomarker may include insulin-like growth factor binding protein 2 (IBP2_HUMAN, UniProt ID P18065). The biomarker may include phosphatidylcholine-sterol acyltransferase (LCAT_HUMAN, UniProt ID P04180). The biomarker may include polymeric immunoglobulin receptor (PIGR_HUMAN, UniProt ID P01833). The biomarker may include tenascin-X (TENX_HUMAN, UniProt ID P22105). The biomarker may include attractin (ATRN_HUMAN, UniProt ID O75882). The biomarker may include intelectin-1 (ITLN1_HUMAN, UniProt ID Q8WWA0). The biomarker may include integrin beta-1 (ITB1_HUMAN, UniProt ID P05556). The biomarker may include immunoglobulin heavy chain constant gamma 2 (IGHG2_HUMAN, UniProt ID P01859). The biomarker may include alpha-N-acetylglucosaminidase (ANAG_HUMAN, UniProt ID P54802). The biomarker may include hepatocyte growth factor activator (HGFA_HUMAN, UniProt ID Q04756). The biomarker may include beta-Ala-His dipeptidase (CNDP1_HUMAN, UniProt ID Q96KN2). The biomarker may include lumican (LUM_HUMAN, UniProt ID P51884). The biomarker may include neural locus notch homolog protein 2 (NOTC2_HUMAN, UniProt ID Q04721). The biomarker may include synaptophysin-like protein 1 (SYPL1_HUMAN, UniProt ID Q16563).The biomarker may include complement factor H-related protein 1 (FHR1_HUMAN, UniProt ID Q03591). The biomarker may include coagulation factor VII (FA7_HUMAN, UniProt ID P08709). The biomarker may include extracellular matrix protein 1 (ECM1_HUMAN, UniProt ID Q16610). The biomarker may include GDH / 6PGL endoplasmic reticulum bifunctional protein (G6PE_HUMAN, UniProt ID O95479). Fragments of any of these proteins may be used. Any of these biomarkers, alone or in combination, may be useful for assessing a lung nodule (e.g., determining the likelihood that the lung nodule is cancerous). Protein measurements can be obtained by use of an internal standard. 【0381】 Some examples of proteins that can be used as biomarkers are shown in Table 15F. One, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, or fifteen of these proteins may be useful as biomarkers, for example, in the evaluation of lung nodules. Any of the following proteins may be useful as such (indicated by their UniProt ID numbers: alpha-2-HS-glycoprotein (FETUA_HUMAN, UniProt ID P02765), fetuin-B (FETUB_HUMAN, UniProt ID Q9UGM5), Src kinase-associated phosphoprotein 2 (SKAP2_HUMAN, UniProt ID O75563), complement C5 (CO5_HUMAN, UniProt ID P01031), collagen alpha-3(VI) chain (CO6A3_HUMAN, UniProt ID P12111), dehydrogenase / reductase SDR family member 7 (DHRS7_HUMAN, UniProt ID Q9Y394), UDP-glucuronate decarboxylase 1 (UXS1_HUMAN, UniProt ID Q8NBZ7-2), complement C1s small component (C1S, A0A087X232), complement C1s small component (C1S_HUMAN, UniProt ID P09871), thrombospondin-1 (TSP1_HUMAN, UniProt ID P07996), tryptophan--tRNA ligase, cytoplasmic (SYWC_HUMAN, UniProt ID P23381), alpha-2-macroglobulin (A2MG_HUMAN, UniProt ID P01023), alpha-actinin-1 (ACTN1_HUMAN, UniProt ID P12814), septin-2 (SEPT2_HUMAN, UniProt ID Q15019-2), apolipoprotein B-100 (APOB_HUMAN, UniProt ID P04114), or complement component C8 beta chain (CO8B_HUMAN, UniProt ID P07358). The biomarker may include alpha-2-HS-glycoprotein (FETUA_HUMAN, UniProt ID P02765).The biomarker may include fetuin-B (FETUB_HUMAN, UniProt ID Q9UGM5). The biomarker may include Src kinase-associated phosphoprotein 2 (SKAP2_HUMAN, UniProt ID O75563). The biomarker may include complement C5 (CO5_HUMAN, UniProt ID P01031). The biomarker may include collagen alpha-3(VI) chain (CO6A3_HUMAN, UniProt ID P12111). The biomarker may include dehydrogenase / reductase SDR family member 7 (DHRS7_HUMAN, UniProt ID Q9Y394). The biomarker may include UDP-glucuronic acid decarboxylase 1 (UXS1_HUMAN, UniProt ID Q8NBZ7) (e.g., Q8NBZ7-2). The biomarker may include complement C1s small component (C1S, A0A087X232). The biomarker may include complement C1s small component (C1S_HUMAN, UniProt ID P09871). The biomarker may include thrombospondin-1 (TSP1_HUMAN, UniProt ID P07996). The biomarker may include tryptophan--tRNA ligase, cytoplasmic (SYWC_HUMAN, UniProt ID P23381). The biomarker may include alpha-2-macroglobulin (A2MG_HUMAN, UniProt ID P01023). The biomarker may include alpha-actinin-1 (ACTN1_HUMAN, UniProt ID P12814). The biomarker may include septin-2 (SEPT2_HUMAN, UniProt ID Q15019) (e.g., Q15019-2). The biomarker may include apolipoprotein B-100 (APOB_HUMAN, UniProt ID P04114). The biomarker may include complement component C8 beta chain (CO8B_HUMAN, UniProt ID P07358). Fragments of any of these proteins may be used.Any of these biomarkers, alone or in combination, may be useful for assessing a lung nodule (e.g., determining the likelihood that a lung nodule is cancerous). In some cases, any of these peptides may be useful as a biomarker when measured after adsorption to particles from a biological fluid sample. 【0382】 Several proteins can be used as biomarkers. One, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, or fifteen of these proteins may be useful as biomarkers in the evaluation of lung cancer, such as non-small cell lung cancer. Any of the following proteins may be useful as such (indicated by their UniProt ID numbers: Sushi, von Willebrand factor type A, EGF and pentraxin domain-containing protein 1 (SVEP1_HUMAN, UniProt ID Q4LDE5), polymeric immunoglobulin receptor (PIGR_HUMAN, UniProt ID P01833), band 3 anion transport protein (B3AT_HUMAN, UniProt ID P02730), cysteine-rich protein 1 (CRIP1_HUMAN, UniProt ID P50238), fibrinogen-like protein 1 (FGL1_HUMAN, UniProt ID Q08830), Cas scaffolding protein family member 4 (CASS4_HUMAN, UniProt ID Q9NQ75), complement component C9 (CO9_HUMAN, UniProt ID P02748), plasminogen activator inhibitor 1 (PAI1_HUMAN, UniProt ID P05121), alpha-1-acid glycoprotein 1 (A1AG1_HUMAN, UniProt ID P02763), apolipoprotein B-100 (APOB_HUMAN, UniProt ID P04114), leukotriene A-4 hydrolase (LKHA4_HUMAN, UniProt ID P09960), beta-Ala-His dipeptidase (CNDP1_HUMAN, UniProt ID Q96KN2), histone H4 (H4_HUMAN, UniProt ID P62805), bone morphogenetic protein 1 (BMP1_HUMAN, UniProt ID P13497), beta-enolase (ENOB_HUMAN, UniProt ID P13929), histone H2A 2-C type (H2A2C_HUMAN, UniProt ID Q16777), protein S100-A12 (S10AC_HUMAN, UniProt IDP80511), insulin-like growth factor-binding protein 2 (IBP2_HUMAN, UniProt ID P18065), protein S100-A8 (S10A8_HUMAN, UniProt ID P05109), complement C4-B (CO4B_HUMAN, UniProt ID P0C0L5), plekstrin (PLEK_HUMAN, UniProt ID P08567), adipocyte plasma membrane-bound protein (APMAP_HUMAN, UniProt ID Q9HDC9), apolipoprotein (a) (APOA_HUMAN, UniProt ID P08519), integrin-binding protein kinase (ILK_HUMAN, UniProt ID Q13418), cytoplasmic dynein 1 heavy chain 1 (DYHC1_HUMAN, UniProt ID Q14204), myosin light chain 12A (J3QRS3_HUMAN, UniProt ID J3QRS3), hepcidin (HEPC_HUMAN, UniProt ID P81172), transforming growth factor-beta-1-induced transcript 1 protein (TGFI1_HUMAN, UniProt ID O43294), latent transforming growth factor-beta binding protein 2 (LTBP2_HUMAN, UniProt ID Q14767), activated RNA polymerase II transcription coactivator p15 (TCP4_HUMAN, UniProt ID P53999), alpha-2-macroglobulin (A2MG_HUMAN, UniProt ID P01023), apolipoprotein A-IV (APOA4_HUMAN, UniProt ID P06727), ribonuclease inhibitor (RINI_HUMAN, UniProt ID P13489), neutrophil defensin 1 (DEF1_HUMAN, UniProt ID P59665), C-X-C motif chemokine 17 (CXL17_HUMAN, UniProt ID Q6UXB2), histone H1.4 (H14_HUMAN, UniProt ID P10412), protein disulfide-isomerase A3 (PDIA3_HUMAN, UniProt ID P30101), PDZ and LIM domain protein 1 (PDLI1_HUMAN, UniProt ID O00151), alpha-actinin-1 (ACTN1_HUMAN, UniProt IDP12814), Serum amyloid A-1 protein (SAA1_HUMAN, UniProt ID P0DJI8), Desmocollin-1 (DSC1_HUMAN, UniProt ID Q08554), Coagulation factor V (FA5_HUMAN, UniProt ID P12259), Alpha-1-antichymotrypsin (AACT_HUMAN, UniProt ID P01011), Myosin-9 (MYH9_HUMAN, UniProt ID P35579), Basement membrane-specific heparan sulfate proteoglycan core protein (PGBM_HUMAN, UniProt ID P98160), Tyrosine-protein kinase SYK (KSYK_HUMAN, UniProt ID P43405), Fibrinogen alpha chain (FIBA_HUMAN, UniProt ID P02671), Tubulin beta-1 chain (TBB1_HUMAN, UniProt ID Q9H4B7), Heparin cofactor 2 (HEP2_HUMAN, UniProt ID P05546), Apolipoprotein A-I (APOA1_HUMAN, UniProt ID P02647), Complement C3 (CO3_HUMAN, UniProt ID P01024), Tropomodulin-3 (TMOD3_HUMAN, UniProt ID Q9NYL9), High-mobility group protein B2 (HMGB2_HUMAN, UniProt ID P26583), Pulmonary surfactant-associated protein B (PSPB_HUMAN, UniProt ID P07988), Interleukin enhancer-binding factor 2 (ILF2_HUMAN, UniProt ID Q12905), Serpin H1 (SERPH_HUMAN, UniProt ID P50454), Reelin (J3KQ66_HUMAN, UniProt ID J3KQ66), WD repeat-containing protein 1 (WDR1_HUMAN, UniProt ID), Flavin reductase (BLVRB_HUMAN, UniProt ID P30043), Inter-alpha-trypsin inhibitor heavy chain H1 (ITIH1_HUMAN, UniProt ID P19827), Glyceraldehyde-3-phosphate dehydrogenase (G3P_HUMAN, UniProt IDP04406), stomachin-like protein 2, mitochondrial (STML2_HUMAN, UniProt ID Q9UJZ1), asporin (ASPN_HUMAN, UniProt ID Q9BXN1), leucine-rich repeat-containing protein 47 (LRC47_HUMAN, UniProt ID Q8N1G4), POSTN-5_HUMAN, SMD3-2_HUMAN, FINC-1_HUMAN, MASP1-2_HUMAN, AMD-3_HUMAN, ILF3-7_HUMAN, VINC-2_HUMAN, ITIH3-2_HUMAN, or FLNA-2_HUMAN. The biomarker may include sushi, von Willebrand factor type A, EGF, and pentraxin domain-containing protein 1 (SVEP1_HUMAN, UniProt ID Q4LDE5). The biomarker may include the polymeric immunoglobulin receptor (PIGR_HUMAN, UniProt ID P01833). The biomarker may include the band 3 anion transport protein (B3AT_HUMAN, UniProt ID P02730). The biomarker may include the cysteine-rich protein 1 (CRIP1_HUMAN, UniProt ID P50238). The biomarker may include fibrinogen-like protein 1 (FGL1_HUMAN, UniProt ID Q08830). The biomarker may include Cas scaffolding protein family member 4 (CASS4_HUMAN, UniProt ID Q9NQ75). The biomarker may include complement component C9 (CO9_HUMAN, UniProt ID P02748). The biomarker may include plasminogen activator inhibitor 1 (PAI1_HUMAN, UniProt ID P05121). The biomarker may include alpha-1-acid glycoprotein 1 (A1AG1_HUMAN, UniProt ID P02763). The biomarker may include apolipoprotein B-100 (APOB_HUMAN, UniProt ID P04114). The biomarker may include leukotriene A-4 hydrolase (LKHA4_HUMAN, UniProt IDIt may include P09960). The biomarker may include beta-Ala-His dipeptidase (CNDP1_HUMAN, UniProt ID Q96KN2). The biomarker may include histone H4 (H4_HUMAN, UniProt ID P62805). The biomarker may include bone morphogenetic protein 1 (BMP1_HUMAN, UniProt ID P13497). The biomarker may include beta-enolase (ENOB_HUMAN, UniProt ID P13929). The biomarker may include histone H2A 2-C type (H2A2C_HUMAN, UniProt ID Q16777). The biomarker may include protein S100-A12 (S10AC_HUMAN, UniProt ID P80511). The biomarker may include insulin-like growth factor binding protein 2 (IBP2_HUMAN, UniProt ID P18065). The biomarker may include protein S100-A8 (S10A8_HUMAN, UniProt ID P05109). The biomarker may include complement C4-B (CO4B_HUMAN, UniProt ID P0C0L5). The biomarker may include plekstrin (PLEK_HUMAN, UniProt ID P08567). The biomarker may include adipocyte plasma membrane-bound protein (APMAP_HUMAN, UniProt ID Q9HDC9). The biomarker may include apolipoprotein (a) (APOA_HUMAN, UniProt ID P08519). The biomarker may include integrin-binding protein kinase (ILK_HUMAN, UniProt ID Q13418). The biomarker may include cytoplasmic dynein 1 heavy chain 1 (DYHC1_HUMAN, UniProt ID Q14204). The biomarker may include myosin light chain 12A (J3QRS3_HUMAN, UniProt ID J3QRS3). The biomarker may include hepcidin (HEPC_HUMAN, UniProt IDIt may include (P81172). The biomarker may include transforming growth factor beta-1-induced transcript 1 protein (TGFI1_HUMAN, UniProt ID O43294). The biomarker may include latent transforming growth factor beta-binding protein 2 (LTBP2_HUMAN, UniProt ID Q14767). The biomarker may include activated RNA polymerase II transcription coactivator p15 (TCP4_HUMAN, UniProt ID P53999). The biomarker may include alpha-2-macroglobulin ( It may include A2MG_HUMAN, UniProt ID P01023. The biomarker may include apolipoprotein A-IV (APOA4_HUMAN, UniProt ID P06727). The biomarker may include ribonuclease inhibitor (RINI_HUMAN, UniProt ID P13489). The biomarker may include neutrophil defensin 1 (DEF1_HUMAN, UniProt ID P59665). The biomarker may include C-X-C motif chemokine 17 (CXL17_HUMAN, UniProt ID Q6UXB2). The biomarker may include histone H1.4 (H14_HUMAN, UniProt ID P10412). The biomarker may include protein disulfide-isomerase A3 (PDIA3_HUMAN, UniProt ID P30101). The biomarker may include PDZ and LIM domain protein 1 (PDLI1_HUMAN, UniProt ID O00151). The biomarker may include alpha-actinin-1 (ACTN1_HUMAN, UniProt ID P12814). The biomarker may include serum amyloid A-1 protein (SAA1_HUMAN, UniProt ID P0DJI8). The biomarker may include desmocollin-1 (DSC1_HUMAN, UniProt ID Q08554). The biomarker may include coagulation factor V (FA5_HUMAN, UniProt ID P12259). The biomarker may include alpha-1-antichymotrypsin (AACT_HUMAN, UniProt ID P01011). The biomarker may include myosin-9 (MYH9_HUMAN, UniProt ID P35579). The biomarker may include basement membrane-specific heparan sulfate proteoglycan core protein (PGBM_HUMAN, UniProt ID P98160). The biomarker may include tyrosine-protein kinase SYK (KSYK_HUMAN, UniProt ID P43405).The biomarker may include fibrinogen alpha chain (FIBA_HUMAN, UniProt ID P02671). The biomarker may include tubulin beta-1 chain (TBB1_HUMAN, UniProt ID Q9H4B7). The biomarker may include heparin cofactor 2 (HEP2_HUMAN, UniProt ID P05546). The biomarker may include apolipoprotein A-I (APOA1_HUMAN, UniProt ID P02647). The biomarker may include complement C3 (CO3_HUMAN, UniProt ID P01024). The biomarker may include tropomodulin-3 (TMOD3_HUMAN, UniProt ID Q9NYL9). The biomarker may include high-mobility group protein B2 (HMGB2_HUMAN, UniProt ID P26583). The biomarker may include pulmonary surfactant-associated protein B (PSPB_HUMAN, UniProt ID P07988). The biomarker may include interleukin enhancer-binding factor 2 (ILF2_HUMAN, UniProt ID Q12905). The biomarker may include serpin H1 (SERPH_HUMAN, UniProt ID P50454). The biomarker may include reelin (J3KQ66_HUMAN, UniProt ID J3KQ66). The biomarker may include WD repeat-containing protein 1 (WDR1_HUMAN, UniProt ID). The biomarker may include flavin reductase (BLVRB_HUMAN, UniProt ID P30043). The biomarker may include inter-alpha-trypsin inhibitor heavy chain H1 (ITIH1_HUMAN, UniProt ID P19827). The biomarker may include glyceraldehyde-3-phosphate dehydrogenase (G3P_HUMAN, UniProt ID P04406). The biomarker may include stomatin-like protein 2, mitochondrial form (STML2_HUMAN, UniProt ID Q9UJZ1).The biomarker may include asporin (ASPN_HUMAN, UniProt ID Q9BXN1). The biomarker may include leucine-rich repeat-containing protein 47 (LRC47_HUMAN, UniProt ID Q8N1G4). The biomarker may include POSTN-5_HUMAN. The biomarker may include SMD3-2_HUMAN. The biomarker may include FINC-1_HUMAN. The biomarker may include MASP1-2_HUMAN. The biomarker may include AMD-3_HUMAN. The biomarker may include ILF3-7_HUMAN. The biomarker may include VINC-2_HUMAN. The biomarker may include ITIH3-2_HUMAN. The biomarker may include FLNA-2_HUMAN. Fragments of any of these proteins may be used. Any of these biomarkers may be useful, alone or in combination, for assessing lung cancer (e.g., non-small cell lung cancer). In some cases, any of these peptides may be useful as biomarkers when measured after adsorption to particles from a biological fluid sample. 【0383】 Any of the following protein biomarkers may be useful for detecting, identifying, or assessing the presence, absence, or likelihood of cancer: SVEP1, PIGR, B3AT,CRIP1,FGL1,CASS4,CO9,PAI1,A1AG1,APOB,LKHA4,CNDP1,H4,BMP1,ENOB,H2A2C,S10AC,IBP2,S10A8,CO4B,PLEK,APMAP,APOA,ILK,DYHC1,J3QRS3,HEPC,TGFI1,LTBP2,TCP4,A2MG,APOA4,RINI,DEF1,CXL17,H14,PDIA3,PDLI1,ACTN1,SAA1,DSC1,FA5,AACT,MYH9,PGBM,KSYK,FIBA,TBB1,HEP2,APOA1,CO3,TMOD3,HMGB2,PSPB,ILF2,SERPH,J3KQ66,WDR1,BLVRB,ITIH1,G3P,STML2,ASPN,LRC47,POSTN-5,SMD3-2,FINC-1,MASP1-2,AMD-3,ILF3-7,VINC-2,ITIH3-2, or FLNA-2.Any of the following protein biomarkers may be useful for detecting, identifying, or assessing the presence, absence, or likelihood of cancer: SVEP1_HUMAN, PIGR_HUMAN, B3AT_HUMAN,CRIP1_HUMAN, FGL1_HUMAN, CASS4_HUMAN, CO9_HUMAN, PAI1_HUMAN, A1AG1_HUMAN, APOB_HUMAN, LKHA4_HUMAN, CNDP1_HUMAN, H4_HUMAN, BMP1_HUMAN, ENOB_HUMAN, H2A2C_HUMAN, S10AC_HUMAN, IBP2_HUMAN, S10A8_HUMAN, CO4B_HUMAN, PLEK_HUMAN, APMAP_HUMAN, APOA_HUMAN, ILK_HUMAN, DYHC1_HUMAN, J3QRS3_HUMAN, HEPC_HUMAN, TGFI1_HUMAN, LTBP2_HUMAN, TCP4_HUMAN, A2MG_HUMAN, APOA4_HUMAN, RINI_HUMAN, DEF1_HUMAN, CXL17_HUMAN, H14_HUMAN, PDIA3_HUMAN, PDLI1_HUMAN, ACTN1_HUMAN, SAA1_HUMAN, DSC1_HUMAN, FA5_HUMAN, AACT_HUMAN, MYH9_HUMAN, PGBM_HUMAN, KSYK_HUMAN, FIBA_HUMAN, TBB1_HUMAN, HEP2_HUMAN, APOA1_HUMAN, CO3_HUMAN, TMOD3_HUMAN, HMGB2_HUMAN, PSPB_HUMAN, ILF2_HUMAN, SERPH_HUMAN, J3KQ66_HUMAN, WDR1_HUMAN, BLVRB_HUMAN, ITIH1_HUMAN, G3P_HUMAN, STML2_HUMAN, ASPN_HUMAN, LRC47_HUMAN, POSTN-5_HUMAN, SMD3-2_HUMAN, FINC-1_HUMAN, MASP1-2_HUMAN, AMD-3_HUMAN, ILF3-7_HUMAN, VINC-2_HUMAN, ITIH3-2_HUMAN, or FLNA-2_HUMAN. The cancer may include lung cancer such as NSCLC. Any number or combination of the aforementioned biomarkers may be used.For example, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, or more of said proteins may be used. 【0384】 Proteomics data can include peptide information such as peptide measurements in biological fluids. Some examples of peptide biomarkers that may be useful in the methods disclosed herein, such as evaluating cancer such as pancreatic cancer, are included in FIG. 140E. Protein measurements can be obtained by use of particles such as those described herein. Any combination or number of such biomarkers can be included. In some cases, a biomarker is useful if its feature importance score exceeds 0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, or 0.10. The feature can include any of the following peptides (shown using one-letter amino acid codes): TELVEPTEYLVVHLK (SEQ ID NO: 1), TFVIIPELVLPNR (SEQ ID NO: 2), LQELHLSSNGLESLSPEFLRPVPQLR (SEQ ID NO: 3), ITLLSALVETR (SEQ ID NO: 4), VVATTQMQAADAR (SEQ ID NO: 5), TFVIIPELVLPNR (SEQ ID NO: 6), LQHLENELTHDIITK (SEQ ID NO: 7), FLENEDRR (SEQ ID NO: 8), LWYENPGVFSPAQLTQIK (SEQ ID NO: 9), QWMENPNNNPIHPNLR (SEQ ID NO: 10), or LEIYQEDQIHFMCPLAR (SEQ ID NO: 11). The biomarker can include TELVEPTEYLVVHLK (SEQ ID NO: 1). The biomarker can include TFVIIPELVLPNR (SEQ ID NO: 2). The biomarker can include LQELHLSSNGLESLSPEFLRPVPQLR (SEQ ID NO: 3). The biomarker can include ITLLSALVETR (SEQ ID NO: 4). The biomarker can include VVATTQMQAADAR (SEQ ID NO: 5). The biomarker can include TFVIIPELVLPNR (SEQ ID NO: 6). The biomarker can include LQHLENELTHDIITK (SEQ ID NO: 7). The biomarker can include FLENEDRR (SEQ ID NO: 8). The biomarker can include LWYENPGVFSPAQLTQIK (SEQ ID NO: 9). The biomarker can include QWMENPNNNPIHPNLR (SEQ ID NO: 10). The biomarker can include LEIYQEDQIHFMCPLAR (SEQ ID NO: 11).Fragments of any of these peptides can be used. Any of these biomarkers can be useful alone or in combination for evaluating cancers such as pancreatic cancer (e.g., for determining the likelihood that a subject has cancer). In some cases, any of these peptides can be useful as a biomarker when measured in conjunction with an internal standard. 【0385】 Some examples of peptides that can be used as biomarkers are shown in Table 15E. One, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, or twenty of these peptides can be useful as biomarkers, for example, in the evaluation of lung nodules. Any of the following peptides can be useful as such (shown using one-letter amino acid codes): STVLTIPEIIIK (SEQ ID NO: 12), TLAFPLTIR (SEQ ID NO: 13), LIQGAPTIR (SEQ ID NO: 14), SSGLVSNAPGVQIR (SEQ ID NO: 15), DGSFSVVITGLR (SEQ ID NO: 16), LGPISADSTTAPLEK (SEQ ID NO: 17), SEAACLAAGPGIR (SEQ ID NO: 18), TDTGFLQTLGHNLFGIYQK (SEQ ID NO: 19), LKPEDITQIQPQQLVLR (SEQ ID NO: 20), GLPAPIEK (SEQ ID NO: 21), LLGPGPAADFSVSVER (SEQ ID NO: 22), YEYLEGGDR (SEQ ID NO: 23), HLEDVFSK (SEQ ID NO: 24), ILGPLSYSK (SEQ ID NO: 25), NCQTVLAPCSPNPCENAAVCK (SEQ ID NO: 26), TVTATFGYPFR (SEQ ID NO: 27), STDTSCVNPPTVQNAHILSR (SEQ ID NO: 28), FSLVSGWGQLLDR (SEQ ID NO: 29), ELLALIQLER (SEQ ID NO: 30), or DAHSVLLSHIFHGR (SEQ ID NO: 31), or fragments thereof. The biomarker can include STVLTIPEIIIK (SEQ ID NO: 12). The biomarker can include TLAFPLTIR (SEQ ID NO: 13). The biomarker can include LIQGAPTIR (SEQ ID NO: 14). The biomarker can include SSGLVSNAPGVQIR (SEQ ID NO: 15). The biomarker can include DGSFSVVITGLR (SEQ ID NO: 16). The biomarker can include LGPISADSTTAPLEK (SEQ ID NO: 17). The biomarker can include SEAACLAAGPGIR (SEQ ID NO: 18). The biomarker can include TDTGFLQTLGHNLFGIYQK (SEQ ID NO: 19). The biomarker can include LKPEDITQIQPQQLVLR (SEQ ID NO: 20).The biomarker may include GLPAPIEK (SEQ ID NO: 21). The biomarker may include LLGPGPAADFSVSVER (SEQ ID NO: 22). The biomarker may include YEYLEGGDR (SEQ ID NO: 23). The biomarker may include HLEDVFSK (SEQ ID NO: 24). The biomarker may include ILGPLSYSK (SEQ ID NO: 25). The biomarker may include NCQTVLAPCSPNPCENAAVCK (SEQ ID NO: 26). The biomarker may include TVTATFGYPFR (SEQ ID NO: 27). The biomarker may include STDTSCVNPPTVQNAHILSR (SEQ ID NO: 28). The biomarker may include FSLVSGWGQLLDR (SEQ ID NO: 29). The biomarker may include ELLALIQLER (SEQ ID NO: 30). The biomarker may include DAHSVLLSHIFHGR (SEQ ID NO: 31). Fragments of any of these peptides may be used. Any of these biomarkers may be useful alone or in combination to evaluate a pulmonary nodule (e.g., to determine the likelihood that the pulmonary nodule is cancerous). In some cases, any of these peptides may be useful as a biomarker when measured in conjunction with an internal standard. 【0386】 The biomarker may include STVLTIPEIIIK (SEQ ID NO: 12) and may be associated with coagulation factor XIII A chain (F13A_HUMAN, UniProt ID P00488). The biomarker may include TLAFPLTIR (SEQ ID NO: 13) and may be associated with endothelial protein C receptor (EPCR_HUMAN, UniProt ID Q9UNN8). The biomarker may include LIQGAPTIR (SEQ ID NO: 14) and may be associated with insulin-like growth factor binding protein 2 (IBP2_HUMAN, UniProt ID P18065). The biomarker may include SSGLVSNAPGVQIR (SEQ ID NO: 15) and may be associated with phosphatidylcholine-sterol acyltransferase (LCAT_HUMAN, UniProt ID P04180). The biomarker may include DGSFSVVITGLR (SEQ ID NO: 16) and may be associated with polymeric immunoglobulin receptor (PIGR_HUMAN, UniProt ID P01833). The biomarker may include LGPISADSTTAPLEK (SEQ ID NO: 17) and may be associated with tenascin-X (TENX_HUMAN, UniProt ID P22105). The biomarker may include SEAACLAAGPGIR (SEQ ID NO: 18) and may be associated with attractin (ATRN_HUMAN, UniProt ID O75882). The biomarker may include TDTGFLQTLGHNLFGIYQK (SEQ ID NO: 19) and may be associated with intelectin-1 (ITLN1_HUMAN, UniProt ID Q8WWA0). The biomarker may include LKPEDITQIQPQQLVLR (SEQ ID NO: 20) and may be associated with integrin beta-1 (ITB1_HUMAN, UniProt P05556). The biomarker may include GLPAPIEK (SEQ ID NO: 21) and may be associated with immunoglobulin heavy chain constant gamma 2 (IGHG2_HUMAN, UniProt P01859).The biomarker may include LLGPGPAADFSVSVER (SEQ ID NO: 22) and may be associated with alpha-N-acetylglucosaminidase (ANAG_HUMAN, UniProt P54802). The biomarker may include YEYLEGGDR (SEQ ID NO: 23) and may be associated with hepatocyte growth factor activator (HGFA_HUMAN, UniProt Q04756). The biomarker may include HLEDVFSK (SEQ ID NO: 24) and may be associated with beta-Ala-His dipeptidase (CNDP1_HUMAN, UniProt Q96KN2). The biomarker may include ILGPLSYSK (SEQ ID NO: 25) and may be associated with lumican (LUM_HUMAN, UniProt P51884). The biomarker may include NCQTVLAPCSPNPCENAAVCK (SEQ ID NO: 26) and may be associated with neural locus notch homolog protein 2 (NOTC2_HUMAN, UniProt Q04721). The biomarker may include TVTATFGYPFR (SEQ ID NO: 27) and may be associated with synaptophysin-like protein 1 (SYPL1_HUMAN, UniProt Q16563). The biomarker may include STDTSCVNPPTVQNAHILSR (SEQ ID NO: 28) and may be associated with complement factor H-related protein 1 (FHR1_HUMAN, UniProt Q03591). The biomarker may include FSLVSGWGQLLDR (SEQ ID NO: 29) and may be associated with coagulation factor VII (FA7_HUMAN, UniProt P08709). The biomarker may include ELLALIQLER (SEQ ID NO: 30) and may be associated with extracellular matrix protein 1 (ECM1_HUMAN, UniProt Q16610). The biomarker may include DAHSVLLSHIFHGR (SEQ ID NO: 31) and may be associated with GDH / 6PGL endoplasmic reticulum bifunctional protein (G6PE_HUMAN, UniProt O95479). Fragments of any of these peptides may be used.Any of the foregoing peptide or protein biomarkers (or combinations of said biomarkers) may be useful for identifying the presence, absence, or likelihood of cancer as described herein. 【0387】 Some examples of peptides that can be used as biomarkers are shown in Table 15F. One, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, or twenty of these peptides can be useful as biomarkers, for example, in the evaluation of lung nodules. Any of the following peptides can be useful as such (shown using one-letter amino acid codes): EHAVEGDCDFQLLK (SEQ ID NO: 32), SQASSCSLQSSDSVPVGLCK (SEQ ID NO: 33), GEFAIDGYSVR (SEQ ID NO: 34), ALVEGVDQLFTDYQIK (SEQ ID NO: 35), LLPYIVGVAQR (SEQ ID NO: 36), HTLNQIDEVK (SEQ ID NO: 37), IDILVNNGGMSQR (SEQ ID NO: 38), LMMDGHEVTVVDNFFTGR (SEQ ID NO: 39), MYGEILSPNYPQAYPSEVEK (SEQ ID NO: 40), NNEEWTVDSCTECHCQNSVTICK (SEQ ID NO: 41), IDTQDIEASHYR (SEQ ID NO: 42), TFIFSDLDYMGMSSGFYK (SEQ ID NO: 43), PDAELSASSVYNLLPEK (SEQ ID NO: 44), ASIHEAWTDGK (SEQ ID NO: 45), LYPWGVVEVENPEHNDFLK (SEQ ID NO: 46), YHWEHTGLTLR (SEQ ID NO: 47), or IGGAIEEVYVSLGVSVGK (SEQ ID NO: 48), or fragments thereof. The biomarker can include EHAVEGDCDFQLLK (SEQ ID NO: 32). The biomarker can include SQASSCSLQSSDSVPVGLCK (SEQ ID NO: 33). The biomarker can include GEFAIDGYSVR (SEQ ID NO: 34). The biomarker can include ALVEGVDQLFTDYQIK (SEQ ID NO: 35). The biomarker can include LLPYIVGVAQR (SEQ ID NO: 36). The biomarker can include HTLNQIDEVK (SEQ ID NO: 37). The biomarker can include IDILVNNGGMSQR (SEQ ID NO: 38). The biomarker can include LMMDGHEVTVVDNFFTGR (SEQ ID NO: 39). The biomarker can include MYGEILSPNYPQAYPSEVEK (SEQ ID NO: 40).The biomarker may include NNEEWTVDSCTECHCQNSVTICK (SEQ ID NO: 41). The biomarker may include IDTQDIEASHYR (SEQ ID NO: 42). The biomarker may include TFIFSDLDYMGMSSGFYK (SEQ ID NO: 43). The biomarker may include PDAELSASSVYNLLPEK (SEQ ID NO: 44). The biomarker may include ASIHEAWTDGK (SEQ ID NO: 45). The biomarker may include LYPWGVVEVENPEHNDFLK (SEQ ID NO: 46). The biomarker may include YHWEHTGLTLR (SEQ ID NO: 47). The biomarker may include IGGAIEEVYVSLGVSVGK (SEQ ID NO: 48). Fragments of any of these peptides may be used. Any of these biomarkers may be useful alone or in combination to evaluate a lung nodule (e.g., to determine the likelihood that the lung nodule is cancerous). In some cases, any of these peptides may be useful as a biomarker when measured after adsorption to particles from a biological fluid sample. 【0388】 The biomarker may contain EHAVEGDCDFQLLK (SEQ ID NO: 32) and may be associated with alpha-2-HS-glycoprotein (FETUA_HUMAN, UniProt ID P02765). The biomarker may contain SQASSCSLQSSDSVPVGLCK (SEQ ID NO: 33) and may be associated with fetuin-B (FETUB_HUMAN, UniProt ID Q9UGM5). The biomarker may contain GEFAIDGYSVR (SEQ ID NO: 34) and may be associated with Src kinase-associated phosphoprotein 2 (SKAP2_HUMAN, UniProt ID O75563). The biomarker may contain ALVEGVDQLFTDYQIK (SEQ ID NO: 35) and may be associated with complement C5 (CO5_HUMAN, UniProt ID P01031). The biomarker may contain LLPYIVGVAQR (SEQ ID NO: 36) and may be associated with collagen alpha-3(VI) chain (CO6A3_HUMAN, UniProt ID P12111). The biomarker may contain HTLNQIDEVK (SEQ ID NO: 37) and may be associated with alpha-2-HS-glycoprotein (FETUA_HUMAN, UniProt ID P02765). The biomarker may contain IDILVNNGGMSQR (SEQ ID NO: 38) and may be associated with dehydrogenase / reductase SDR family member 7 (DHRS7_HUMAN, UniProt ID Q9Y394). The biomarker may contain LMMDGHEVTVVDNFFTGR (SEQ ID NO: 39) and may be associated with UDP-glucuronate decarboxylase 1 (UXS1_HUMAN, UniProt ID Q8NBZ7-2). The biomarker may contain MYGEILSPNYPQAYPSEVEK (SEQ ID NO: 40) and may be associated with complement C1s small component (UniProt ID A0A087X232). The biomarker may contain NNEEWTVDSCTECHCQNSVTICK (SEQ ID NO: 41) and may be associated with thrombospondin-1 (TSP1_HUMAN, UniProt ID P07996).The biomarker may include IDTQDIEASHYR (SEQ ID NO: 42) and may be associated with complement C5 (CO5_HUMAN, UniProt ID P01031). The biomarker may include TFIFSDLDYMGMSSGFYK (SEQ ID NO: 43) and may be associated with tryptophan-tRNA ligase, cytoplasmic (SYWC_HUMAN, UniProt ID P23381). The biomarker may include PDAELSASSVYNLLPEK (SEQ ID NO: 44) and may be associated with alpha-2-macroglobulin (A2MG_HUMAN, UniProt ID P01023). The biomarker may include ASIHEAWTDGK (SEQ ID NO: 45) and may be associated with alpha-actinin-1 (ACTN1_HUMAN, UniProt ID P12814). The biomarker may include LYPWGVVEVENPEHNDFLK (SEQ ID NO: 46) and may be associated with septin-2 (SEPT2_HUMAN, UniProt ID Q15019-2). The biomarker may include YHWEHTGLTLR (SEQ ID NO: 47) and may be associated with apolipoprotein B-100 (APOB_HUMAN, UniProt ID P04114). The biomarker may include IGGAIEEVYVSLGVSVGK (SEQ ID NO: 48) and may be associated with complement component C8 beta chain (CO8B_HUMAN, UniProt ID P07358). Fragments of any of these peptides may be used. Any of the aforementioned peptide or protein biomarkers (or combinations of said biomarkers) may be useful for identifying the presence, absence, or likelihood of the cancers described herein. 【0389】 Proteomics data can include peptide information such as peptide measurements in biological fluids. Some examples of peptide biomarkers that may be useful in the methods disclosed herein, such as evaluating cancers such as lung cancer (e.g., non-small cell lung cancer). Protein measurements can be obtained by use of particles such as those described herein. Any combination or number of such biomarkers can be included. In some cases, a biomarker is useful when its feature importance score exceeds 0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, or 0.10. The features are the following peptides (shown using one-letter amino acid codes): LC(UniMod:4)PSGMYTEYIHSR (SEQ ID NO:49), LCPSGMYTEYIHSR (SEQ ID NO:139), NADLQVLKPEPELVYEDLR (SEQ ID NO:50), ASTPGAAAQIQEVK (SEQ ID NO:51), PYC(UniMod:4)NHPC(UniMod:4)YAAMFGPK (SEQ ID NO:52), PYCNHPCYAAMFGPK (SEQ ID NO:140), QLLQENEVQFLDK (SEQ ID NO:53), AISAFHGSLSSSQPAEIITQSK (SEQ ID NO:54), FEGIAC(UniMod:4)EISK (SEQ ID NO:55), FEGIACEISK (SEQ ID NO:141), FIINDWVK (SEQ ID NO:56), YVGGQEHFAHLLILRDTK (SEQ ID NO:57), SVGFHLPSR (SEQ ID NO:58), GSPMEISLPIALSK (SEQ ID NO:59), M(UniMod:35)VVSMTLGLHPWIANIDDTQYLAAK (SEQ ID NO:60), MVVSMTLGLHPWIANIDDTQYLAAK (SEQ ID NO:142), TVTAM(UniMod:35)DVVYALK (SEQ ID NO:61), TVTAMDVVYALK (SEQ ID NO:143), C(UniMod:4)SC(UniMod:4)DPGYELAPDKR (SEQ ID NO:62), CSCDPGYELAPDKR (SEQ ID NO:144), GNPTVEVDLHTAK (SEQ ID NO:63), HLQLAIRNDEELNK (SEQ ID NO:64), FQDGDLTLYQSNTILR (SEQ ID NO:65), IRPNDFIPNVI (SEQ ID NO:66), TKLEEHLEGIVNIFHQYSVRK (SEQ ID NO:67),GDPEC (UniMod:4) HLFYNEQQEAR (SEQ ID NO: 68), GDPECHLFYNEQQEAR (SEQ ID NO: 145), ALNSIIDVYHK (SEQ ID NO: 69), DDPDAPLQPVTPLQLFEGR (SEQ ID NO: 70), KSEEENLFEIITADEVHYFLQAATPK (SEQ ID NO: 71), FPNGVQLSPAEDFVLVAETTMAR (SEQ ID NO: 72), LYFMHFNLESSYLC (UniMod:4) EYDYVK (SEQ ID NO: 73), LYFMHFNLESSYLCEYDYVK (SEQ ID NO: 146), LFDYC (UniMod:4) DIPLC (UniMod:4) ASSSFDC (UniMod:4) GK (SEQ ID NO: 74), LFDYCDIPLCASSSFDCGK (SEQ ID NO: 147), AEQC (UniMod:4) C (UniMod:4) EETASSISLHGK (SEQ ID NO: 75), AEQCCEETASSISLHGK (SEQ ID NO: 148), VALEGLRPTIPPGISPHVC (UniMod:4) K (SEQ ID NO: 76), VALEGLRPTIPPGISPHVCK (SEQ ID NO: 149), VWEQIDQMK (SEQ ID NO: 77), FTDEEVDELYREAPIDK (SEQ ID NO: 78), DTHFPIC (UniMod:4) IFC (UniMod:4) C (UniMod:4) GC (UniMod:4) C (UniMod:4) HR (SEQ ID NO: 79), DTHFPICIFCCGCCHR (SEQ ID NO: 150), RQDNEILIFWSK (SEQ ID NO: 80), QDNEILIFWSK (SEQ ID NO: 81), EVGTVLSQVYSK (SEQ ID NO: 82), MVTALGTHWHPEHFC (UniMod:4) C (UniMod:4) VSC (UniMod:4) GEPFGDEGFHER (SEQ ID NO: 83), MVTALGTHWHPEHFCCVSCGEPFGDEGFHER (SEQ ID NO: 151), EVTFHC (UniMod:4) HEGYILHGAPK (SEQ ID NO: 84), EVTFHCHEGYILHGAPK (SEQ ID NO: 152), GAGGQSMSEAPTGDHAPAPTR (SEQ ID NO: 85), DGSFSVVITGLR (SEQ ID NO: 86), GISLNPEQWSQLK (SEQ ID NO: 87), LVHVEEPHTETVR (SEQ ID NO: 88), RVEPYGENFNK (SEQ ID NO: 89), LDDC (UniMod:4) GLTEAR (SEQ ID NO: 90),LDDCGLTEAR (SEQ ID NO: 153), LVQAAQMLQSDPYSVPAR (SEQ ID NO: 91), DFLGFYVVDSHR (SEQ ID NO: 92), YGTC (UniMod:4)IYQGR (SEQ ID NO: 93), YGTCIYQGR (SEQ ID NO: 154), WLQEGGQEC (UniMod:4)EC (UniMod:4)K (SEQ ID NO: 94), WLQEGGQECECK (SEQ ID NO: 155), ASGPPVSELITK (SEQ ID NO: 95), ELSDFISYLQR (SEQ ID NO: 96), EGHVLQGPSVLK (SEQ ID NO: 97), MNLASEPQEVLHIGSAHNR (SEQ ID NO: 98), FLILPDMLK (SEQ ID NO: 99), GISQEQMNEFR (SEQ ID NO: 100), DPNHFRPAGLPEK (SEQ ID NO: 101), VPSHLQAETLVGK (SEQ ID NO: 102), NLHFLTTQEDYTLK (SEQ ID NO: 103), SEAYNTFSER (SEQ ID NO: 104), AVLDVFEEGTEASAATAVK (SEQ ID NO: 105), VIQYLAYVASSHK (SEQ ID NO: 106), ASYAQQPAESR (SEQ ID NO: 107), YLEESNFVHR (SEQ ID NO: 108), GSFTYFAPSNEAWDNLDSDIR (SEQ ID NO: 109), ALTDMPQM (UniMod:35)R (SEQ ID NO: 110), LAVNM (UniMod:35)VPFPR (SEQ ID NO: 111), TSC (UniMod:4)LLFMGR (SEQ ID NO: 112), QQQHLFGSNVTDC (UniMod:4)SGNFC (UniMod:4)LFR (SEQ ID NO: 113), ALTDMPQMR (SEQ ID NO: 156), LAVNMVPFPR (SEQ ID NO: 157), TSCLLFMGR (SEQ ID NO: 158), QQQHLFGSNVTDCSGNFCLFR (SEQ ID NO: 159), DYVSQFEGSALGK (SEQ ID NO: 114), DSITTWEILAVSMSDK (SEQ ID NO: 115), FC (UniMod:4)NIMGSSNGVDQEHFSNVVK (SEQ ID NO: 116), FCNIMGSSNGVDQEHFSNVVK (SEQ ID NO: 160), SEHPGLSIGDTAK (SEQ ID NO: 117), QFVEQHTPQLLTLVPR (SEQ ID NO: 118), NQDLAPNSAEQASILSLVTK (SEQ ID NO: 119), TDGALLVNAMFFK (SEQ ID NO: 120)DDFEGQLESDRFLLMSGGK (SEQ ID NO: 121), SIQC (UniMod:4)LTVHK (SEQ ID NO: 122), SIQCLTVHK (SEQ ID NO: 161), EDITQSAQHALR (SEQ ID NO: 123), VVAC (UniMod:4)TSAFLLWDPTK (SEQ ID NO: 124), VVACTSAFLLWDPTK (SEQ ID NO: 162), NYPMHVFAYR (SEQ ID NO: 125), MEEVEAMLLPETLK (SEQ ID NO: 126), ADVQAHGEGQEFSITC (UniMod:4)LVDEEEM (UniMod:35)K (SEQ ID NO: 127), ADVQAHGEGQEFSITCLVDEEEMK (SEQ ID NO: 163), DFALLSLQVPLK (SEQ ID NO: 128), LLIYAVLPTGDVIGDSAK (SEQ ID NO: 129), VDIVAINDPFIDLNYMVYMFQYDSTHGK (SEQ ID NO: 130), AEQINQAAGEASAVLAK (SEQ ID NO: 131), TPAYYPNAGLIK (SEQ ID NO: 132), QGENGQMM (UniMod:35)SC (UniMod:4)TC (UniMod:4)LGNGK (SEQ ID NO: 133), QGENGQMMSCTCLGNGK (SEQ ID NO: 164), YWEMQPATFR (SEQ ID NO: 134), HGEYWLGNK (SEQ ID NO: 135), FVPAEMGTHTVSVK (SEQ ID NO: 136), NALGPGLSPELGPLPALR (SEQ ID NO: 137), or TKLEEHLEGIVNIFHQYSVR (SEQ ID NO: 138) may be included. Fragments of any of these peptides may be used. Any of these biomarkers may be useful alone or in combination to evaluate a lung nodule (e.g., to determine the likelihood that a lung nodule is cancerous). In some cases, any of these peptides may be useful as a biomarker when measured after adsorption to particles from a biological fluid sample. The biomarker may include LC (UniMod:4)PSGMYTEYIHSR (SEQ ID NO: 49). The biomarker may include NADLQVLKPEPELVYEDLR (SEQ ID NO: 50). The biomarker may include ASTPGAAAQIQEVK (SEQ ID NO: 51). The biomarker isIt may include PYC (UniMod:4) NHPC (UniMod:4) YAAMFGPK (SEQ ID NO: 52). The biomarker may include QLLQENEVQFLDK (SEQ ID NO: 53). The biomarker may include AISAFHGSLSSSQPAEIITQSK (SEQ ID NO: 54). The biomarker may include FEGIAC (UniMod:4) EISK (SEQ ID NO: 55). The biomarker may include FIINDWVK (SEQ ID NO: 56). The biomarker may include YVGGQEHFAHLLILRDTK (SEQ ID NO: 57). The biomarker may include SVGFHLPSR (SEQ ID NO: 58). The biomarker may include GSPMEISLPIALSK (SEQ ID NO: 59). The biomarker may include M (UniMod:35) VVSMTLGLHPWIANIDDTQYLAAK (SEQ ID NO: 60). The biomarker may include TVTAM (UniMod:35) DVVYALK (SEQ ID NO: 61). The biomarker may include C (UniMod:4) SC (UniMod:4) DPGYELAPDKR (SEQ ID NO: 62). The biomarker may include GNPTVEVDLHTAK (SEQ ID NO: 63). The biomarker may include HLQLAIRNDEELNK (SEQ ID NO: 64). The biomarker may include FQDGDLTLYQSNTILR (SEQ ID NO: 65). The biomarker may include IRPNDFIPNVI (SEQ ID NO: 66). The biomarker may include TKLEEHLEGIVNIFHQYSVRK (SEQ ID NO: 67). The biomarker may include GDPEC (UniMod:4) HLFYNEQQEAR (SEQ ID NO: 68). The biomarker may include ALNSIIDVYHK (SEQ ID NO: 69). The biomarker may include DDPDAPLQPVTPLQLFEGR (SEQ ID NO: 70). The biomarker may include KSEEENLFEIITADEVHYFLQAATPK (SEQ ID NO: 71). The biomarker may include FPNGVQLSPAEDFVLVAETTMAR (SEQ ID NO: 72). The biomarker may include LYFMHFNLESSYLC (UniMod:4) EYDYVK (SEQ ID NO: 73). The biomarker may includeIt may include LFDYC (UniMod:4) DIPLC (UniMod:4) ASSSFDC (UniMod:4) GK (SEQ ID NO: 74). The biomarker may include AEQC (UniMod:4) C (UniMod:4) EETASSISLHGK (SEQ ID NO: 75). The biomarker may include VALEGLRPTIPPGISPHVC (UniMod:4) K (SEQ ID NO: 76). The biomarker may include VWEQIDQMK (SEQ ID NO: 77). The biomarker may include FTDEEVDELYREAPIDK (SEQ ID NO: 78). Biomarker, The car may include DTHFPIC (UniMod:4) IFC (UniMod:4) C (UniMod:4) GC (UniMod:4) C (UniMod:4) HR (SEQ ID NO: 79). The biomarker may include RQDNEILIFWSK (SEQ ID NO: 80). The biomarker may include QDNEILIFWSK (SEQ ID NO: 81). The biomarker may include EVGTVLSQVYSK (SEQ ID NO: 82). The biomarker may include MVTALGTHWHPEHFC (UniMod:4) C (UniMod:4) VSC (UniMod:4) GEPFGDEGFHER (SEQ ID NO: 83). The biomarker may include EVTFHC (UniMod:4) HEGYILHGAPK (SEQ ID NO: 84). The biomarker may include GAGGQSMSEAPTGDHAPAPTR (SEQ ID NO: 85). The biomarker may include DGSFSVVITGLR (SEQ ID NO: 86). The biomarker may include GISLNPEQWSQLK (SEQ ID NO: 87). The biomarker may include LVHVEEPHTETVR (SEQ ID NO: 88). The biomarker may include RVEPYGENFNK (SEQ ID NO: 89). The biomarker may include LDDC (UniMod:4) GLTEAR (SEQ ID NO: 90). The biomarker may include LVQAAQMLQSDPYSVPAR (SEQ ID NO: 91). The biomarker may include DFLGFYVVDSHR (SEQ ID NO: 92). The biomarker may include YGTC (UniMod:4) IYQGR (SEQ ID NO: 93). The biomarker may include WLQEGGQEC (UniMod:4) EC (UniMod:4) K (SEQ ID NO: 94). The biomarker may include ASGPPVSELITK (SEQ ID NO: 95). The biomarker may include ELSDFISYLQR (SEQ ID NO: 96). The biomarker may include EGHVLQGPSVLK (SEQ ID NO: 97). The biomarker may include MNLASEPQEVLHIGSAHNR (SEQ ID NO: 98). The biomarker may include FLILPDMLK (SEQ ID NO: 99). The biomarker may include GISQEQMNEFR (SEQ ID NO: 100). The biomarker may include DPNHFRPAGLPEK (SEQ ID NO: 101).The biomarker may include VPSHLQAETLVGK (SEQ ID NO: 102). The biomarker may include NLHFLTTQEDYTLK (SEQ ID NO: 103). The biomarker may include SEAYNTFSER (SEQ ID NO: 104). The biomarker may include AVLDVFEEGTEASAATAVK (SEQ ID NO: 105). The biomarker may include VIQYLAYVASSHK (SEQ ID NO: 106). The biomarker may include ASYAQQPAESR (SEQ ID NO: 107). The biomarker may include YLEESNFVHR (SEQ ID NO: 108). The biomarker may include GSFTYFAPSNEAWDNLDSDIR (SEQ ID NO: 109). The biomarker may include ALTDMPQM (UniMod:35)R (SEQ ID NO: 110). The biomarker may include LAVNM (UniMod:35)VPFPR (SEQ ID NO: 111). The biomarker may include TSC (UniMod:4)LLFMGR (SEQ ID NO: 112). The biomarker may include QQQHLFGSNVTDC (UniMod:4)SGNFC (UniMod:4)LFR (SEQ ID NO: 113). The biomarker may include DYVSQFEGSALGK (SEQ ID NO: 114). The biomarker may include DSITTWEILAVSMSDK (SEQ ID NO: 115). The biomarker may include FC (UniMod:4)NIMGSSNGVDQEHFSNVVK (SEQ ID NO: 116). The biomarker may include SEHPGLSIGDTAK (SEQ ID NO: 117). The biomarker may include QFVEQHTPQLLTLVPR (SEQ ID NO: 118). The biomarker may include NQDLAPNSAEQASILSLVTK (SEQ ID NO: 119). The biomarker may include TDGALLVNAMFFK (SEQ ID NO: 120). The biomarker may include DDFEGQLESDRFLLMSGGK (SEQ ID NO: 121). The biomarker may include SIQC (UniMod:4)LTVHK (SEQ ID NO: 122). The biomarker may include EDITQSAQHALR (SEQ ID NO: 123). The biomarker may include VVAC (UniMod:4)TSAFLLWDPTK (SEQ ID NO: 124). The biomarker may include NYPMHVFAYR (SEQ ID NO: 125).The biomarker may include MEEVEAMLLPETLK (SEQ ID NO: 126). The biomarker may include ADVQAHGEGQEFSITC (UniMod:4)LVDEEEM (UniMod:35)K (SEQ ID NO: 127). The biomarker may include DFALLSLQVPLK (SEQ ID NO: 128). The biomarker may include LLIYAVLPTGDVIGDSAK (SEQ ID NO: 129). The biomarker may include VDIVAINDPFIDLNYMVYMFQYDSTHGK (SEQ ID NO: 130). The biomarker may include AEQINQAAGEASAVLAK (SEQ ID NO: 131). The biomarker may include TPAYYPNAGLIK (SEQ ID NO: 132). The biomarker may include QGENGQMM (UniMod:35)SC (UniMod:4)TC (UniMod:4)LGNGK (SEQ ID NO: 133). The biomarker may include YWEMQPATFR (SEQ ID NO: 134). The biomarker may include HGEYWLGNK (SEQ ID NO: 135). The biomarker may include FVPAEMGTHTVSVK (SEQ ID NO: 136). The biomarker may include NALGPGLSPELGPLPALR (SEQ ID NO: 137). The biomarker may include TKLEEHLEGIVNIFHQYSVR (SEQ ID NO: 138). The biomarker may include LCPSGMYTEYIHSR (SEQ ID NO: 139). The biomarker may include PYCNHPCYAAMFGPK (SEQ ID NO: 140). The biomarker may include FEGIACEISK (SEQ ID NO: 141). The biomarker may include MVVSMTLGLHPWIANIDDTQYLAAK (SEQ ID NO: 142). The biomarker may include TVTAMDVVYALK (SEQ ID NO: 143). The biomarker may include CSCDPGYELAPDKR (SEQ ID NO: 144). The biomarker may include GDPECHLFYNEQQEAR (SEQ ID NO: 145). The biomarker may include LYFMHFNLESSYLCEYDYVK (SEQ ID NO: 146). The biomarker may include LFDYCDIPLCASSSFDCGK (SEQ ID NO: 147). The biomarker may include AEQCCEETASSISLHGK (SEQ ID NO: 148).The biomarker may include VALEGLRPTIPPGISPHVCK (SEQ ID NO: 149). The biomarker may include DTHFPICIFCCGCCHR (SEQ ID NO: 150). The biomarker may include MVTALGTHWHPEHFCCVSCGEPFGDEGFHER (SEQ ID NO: 151). The biomarker may include EVTFHCHEGYILHGAPK (SEQ ID NO: 152). The biomarker may include LDDCGLTEAR (SEQ ID NO: 153). The biomarker may include YGTCIYQGR (SEQ ID NO: 154). The biomarker may include WLQEGGQECECK (SEQ ID NO: 155). The biomarker may include ALTDMPQMR (SEQ ID NO: 156). The biomarker may include LAVNMVPFPR (SEQ ID NO: 157). The biomarker may include TSCLLFMGR (SEQ ID NO: 158). The biomarker may include QQQHLFGSNVTDCSGNFCLFR (SEQ ID NO: 159). The biomarker may include and the biomarker may include FCNIMGSSNGVDQEHFSNVVK (SEQ ID NO: 160). The biomarker may include SIQCLTVHK (SEQ ID NO: 161). The biomarker may include VVACTSAFLLWDPTK (SEQ ID NO: 162). The biomarker may include ADVQAHGEGQEFSITCLVDEEEMK (SEQ ID NO: 163). The biomarker may include QGENGQMMSCTCLGNGK (SEQ ID NO: 164). Any (or combination of the foregoing biomarkers) may be useful for identifying the presence, absence, or likelihood of cancer described herein. 【0390】 Some embodiments include peptide transitions. Some embodiments include the use of multiple peptide transitions. For example, measurements of multiple peptide transitions obtained from a biological fluid sample may be useful in a diagnostic method or any multi-omics method. 【0391】 In some embodiments, the multi-omics data includes measurements of more than 10 peptide or protein groups, more than 15 peptide or protein groups, more than 20 peptide or protein groups, more than 25 peptide or protein groups, more than 30 peptide or protein groups, more than 35 peptide or protein groups, more than 40 peptide or protein groups, more than 45 peptide or protein groups, more than 50 peptide or protein groups, more than 75 peptide or protein groups, more than 100 peptide or protein groups, more than 250 peptide or protein groups, more than 500 peptide or protein groups, more than 1,000 peptide or protein groups, more than 2,500 peptide or protein groups, more than 5,000 peptide or protein groups, more than 10,000 peptide or protein groups, more than 15,000 peptide or protein groups, or more than 20,000 peptide or protein groups. In some embodiments, the multi-omics data includes measurements of at least about 10 peptide or protein groups, at least about 15 peptide or protein groups, at least about 20 peptide or protein groups, at least about 25 peptide or protein groups, at least about 30 peptide or protein groups, at least about 35 peptide or protein groups, at least about 40 peptide or protein groups, at least about 45 peptide or protein groups, at least about 50 peptide or protein groups, at least about 75 peptide or protein groups, at least about 100 peptide or protein groups, at least about 250 peptide or protein groups, at least about 500 peptide or protein groups, at least about 1,000 peptide or protein groups, at least about 2,500 peptide or protein groups, at least about 5,000 peptide or protein groups, at least about 10,000 peptide or protein groups, at least about 15,000 peptide or protein groups, or at least about 20,000 peptide or protein groups.In some embodiments, the protein data includes measurements of a peptide or protein group of 10 or less, 15 or less, 20 or less, 25 or less, 30 or less, 35 or less, 40 or less, 45 or less, 50 or less, 75 or less, 100 or less, 250 or less, 500 or less, 1,000 or less, 2,500 or less, 5,000 or less, 10,000 or less, 15,000 or less, or 20,000 or less. The peptide or protein group can comprise or consist of peptides. The peptide or protein group can comprise or consist of protein groups. 【0392】 The protein may also include post-translational modifications (PTMs). Examples of PTMs can include glycosylation. The protein or peptide may include glycoprotein or glycopeptide. The protein may include glycoprotein. The peptide may include glycopeptide. Examples of PTMs can include phosphorylation. The protein or peptide may include phosphoprotein or phosphopeptide. The protein may include phosphoprotein. The peptide may include phosphopeptide. Examples of PTMs can include carboxamidomethylation. The protein or peptide may include carbamidomethyl protein or carbamidomethyl peptide. The protein may include carbamidomethyl protein. The peptide may include carbamidomethyl peptide. Examples of PTMs can include oxidation or hydroxylation. The protein or peptide may include oxidized or hydroxylated protein or oxidized or hydroxylated peptide. The protein may include oxidized protein. The protein may include hydroxylated protein. The peptide may include oxidized peptide. The peptide may include hydroxylated peptide. 【0393】 Proteomics data can be generated by any of a variety of methods. Generation of proteomics data may include the use of detection reagents that bind to a peptide or protein to generate a detectable signal. After using a detection reagent that binds to a peptide or protein to generate a detectable signal, a reading indicating the presence, absence, or amount of the protein or peptide can be obtained. Generation of proteomics data can include concentration, filtration, or centrifugation of the sample. 【0394】 Proteomic data can be generated using mass spectrometry, chromatography, liquid chromatography, high performance liquid chromatography, solid phase chromatography, lateral flow assay, immunoassay, enzyme-linked immunosorbent assay, Western blot, dot blot, or immunostaining, or combinations thereof. Some examples of methods for generating proteomic data include the use of mass spectrometry, protein chips, or reverse phase protein microarrays. Proteomic data can also be generated using immunoassays, such as enzyme-linked immunosorbent assay, Western blot, dot blot, or immunohistochemical assays. The generation of proteomic data can include the use of an immunoassay panel. 【0395】 One way to obtain proteomic data involves the use of mass spectrometry. Examples of mass spectrometry methods include separating proteins in parallel from different samples using high resolution two-dimensional electrophoresis, followed by selecting or staining differentially expressed proteins identified by mass spectrometry. In another method, stable isotope tags are used to differentially label proteins from two different complex mixtures. Proteins within the complex mixtures are isotopically labeled and then digested to obtain labeled peptides. Next, the labeled mixtures are combined and the peptides can be separated by multidimensional liquid chromatography and analyzed by tandem mass spectrometry. Mass spectrometry methods can include the use of liquid chromatography-mass spectrometry (LC-MS), a technique that combines the physical separation capabilities of liquid chromatography (e.g., HPLC) with mass spectrometry. 【0396】 The protein can be concentrated before assay or measurement. In concentration, it is possible to concentrate a certain set of proteins and not another set, or to concentrate a single protein and not another protein. Concentration can be obtained through the use of an affinity reagent, for example, by incubating the affinity reagent with the sample before measuring the protein in the sample. The affinity reagent may include an antibody. The affinity reagent may include particles such as nanoparticles. The protein can be adsorbed to the affinity reagent, separated from the remaining portion of the sample, and then assayed using the proteomics assays described herein. 【0397】 The generation of proteomics data can include contacting a sample with particles such that the particles adsorb biomolecules including the protein. The adsorbed protein can be part of a biomolecular corona. The adsorbed protein can be measured or identified in generating proteomics data. 【0398】 The generation of proteomics data may include the use of a known amount of an internal reference protein. The reference protein can be labeled. The label may include an isotope label. The generation of proteomics data may include the use of a known amount of an isotope-labeled internal reference protein (referred to as "PiQuant"). The internal reference protein may be spiked into the sample. The internal reference protein can be used to identify the mass spectrum of individual endogenous proteins. The internal reference protein can be used as a standard for determining the amount of individual endogenous proteins. Proteomics measurements can be generated based on the amount of protein added to one sample of one or more biological fluid samples. Proteomics measurements can be generated based on the amount of labeled protein added to one sample of one or more biological fluid samples. In some embodiments, the proteomics data can include spatial proteomics data, and the spatial proteomics data can include detecting and quantifying the intracellular localization of proteins in cells. The spatial proteomics data can be obtained by microscopy, mass spectrometry, and machine learning applications for data analysis. In some embodiments, the spatial proteomics data is in situ proteomics data. 【0399】 Transcriptomics data The data such as multi-omics data described in this specification can include transcription data or transcriptomics data. The transcriptomics data can include data on nucleotide transcripts such as RNA. Examples of RNA include messenger RNA (mRNA), ribosomal RNA (rRNA), signal recognition particle (SRP) RNA, transfer RNA (tRNA), small nuclear RNA (snRNA), small nucleolar RNA (snoRNA), long non-coding RNA (lncRNA), microRNA (miRNA), non-coding RNA (ncRNA), or piwi-interacting RNA (piRNA), or combinations thereof. RNA may include mRNA. RNA may include miRNA. Transcriptomics data can be distinguished by subtype, and each subtype contains different types of RNA or transcripts. For example, mRNA data is included in one subtype, and data on one or more types of small non-coding RNAs such as miRNA or piRNA may be included in another subtype. miRNA may include 5p miRNA or 3p miRNA. 【0400】 The transcriptomics data may include information regarding the presence, absence, or amount of various RNAs. For example, the transcriptomics data may include the amount of RNA. The amount of RNA can be indicated as the concentration or number of RNA molecules, e.g., the concentration of RNA in a biological fluid. The amount of RNA can be compared to another RNA or another biomolecule. The transcriptomics data may include information regarding the presence of RNA. The transcriptomics data may include information regarding the absence of RNA. Aspects described in relation to the transcriptomics data may relate to transcripts or RNA data, and vice versa. 【0401】 Transcriptomic data generally includes data on a large number of RNAs. For example, transcriptomic data can include information on the presence, absence, or quantity of 1000 or more RNAs. In some cases, transcriptomic data may include information on the presence, absence, or quantity of 5000, 10,000, 20,000, or more RNAs. Transcriptomic data can further include up to approximately 200,000 transcripts. The transcript quantity can include copy number. Transcriptomic data may include a set of transcripts defined by any of the aforementioned number of RNAs or transcripts. Some examples of mRNAs that can be included in transcriptomic data are shown in FIG. 10B or FIG. 15. Some examples of microRNAs that can be included in transcriptomic data are shown in FIG. 11B or FIG. 15. 【0402】 Some examples of mRNAs that can be used as biomarkers are shown in FIG. 10B. One, two, three, four, five, six, seven, eight, nine, or ten of the mRNAs included in FIG. 10B can be used as biomarkers, for example, when determining whether a lung nodule is cancerous or when determining the likelihood thereof. Some examples of microRNAs that can be used as biomarkers are shown in FIG. 11B. One, two, three, four, five, six, seven, eight, nine, or ten of the microRNAs included in FIG. 11B can be used as biomarkers, for example, when determining whether a lung nodule is cancerous or when determining the likelihood thereof. 【0403】 Some examples of RNAs that can be used as biomarkers are shown in Table 15B, which includes mRNAs. One, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, or twenty of these RNAs can be useful as biomarkers, for example, in the evaluation of lung nodules. Any of the following RNAs can be useful as such (presented as Ensembl reference numbers): ENSG00000224067.2 (Description: Pseudogene similar to part of HLA-B associated transcript 2 (BAT2)), ENSG00000196735.13 (Description: Major histocompatibility complex class II, DQ alpha 1, HLA-DQA1), ENSG00000287647.1 (Description: Antisense to AK5), ENSG00000230797.3 (Description: YY2 transcription factor, YY2), ENSG00000287219.1 (Description: Novel human transcript derived from positions 38,675,876 to 38,677,800 on the forward strand of chromosome 22 in genome build GRCh38), ENSG00000271543.1 (Description: Ribosomal protein L6 (RPL6) pseudogene), ENSG00000223711.1 (Description: AC091633.3 (clone-based (Vega) gene) at positions 195,270,871 to 195,277,400 on the forward strand of chromosome 3 in human genome build GRCh37), ENSG00000223711.2 (Description: Novel transcript at positions 195,543,418 to 195,550,581 on the forward strand of chromosome 3 in genome build GRCh38), ENSG00000177602.5 (Description: Histone H3 related protein kinase, HASPIN), ENSG00000144671.11 (Description: Solute carrier family 22 member 14, SLC22A14), ENSG00000129673.10 (Description: Arylalkylamine N-acetyltransferase, AANAT), ENSG00000265817.4 (Description: Fibrinogen silencer binding protein, FSBP), ENSG00000108924.14 (Description: HLF transcription factor, PAR bZIP family member, HLF), ENSG00000232125.5 (Description: Distroterin, DYTN), ENSG00000252800.1 (Description: A human transcript derived from positions 63,479,272 to 63,479,413 on the forward strand of chromosome 14 in genome build GRCh38, and this RNA biomarker may correspond to small Cajal body-specific RNA 20 (SCARNA20)), ENSG00000287537.1 (Description: A novel human transcript derived from positions 49,536,677 to 49,538,894 on the reverse strand of chromosome 12 in genome build GRCh38), ENSG00000196405.13 (Description: Enah / Vasp-like, EVL), ENSG00000250893.1 (Description: A novel human transcript derived from positions 40,426,119 to 40,427,585 on the forward strand of chromosome 4 in genome build GRCh38), ENSG00000153446.16 (Description: Chromosome 16 open reading frame 89, C16orf89), ENSG00000284630.1 (Description: A novel human transcript derived from positions 21,657,811 to 21,661,021 on the forward strand of chromosome 22 in genome build GRCh38), or ENSG00000284687.1 (Description: An RNA-binding protein, fox-1 homolog (C. elegans) 1 (RBFOX1) human pseudogene, derived from positions 8,390,270 to 8,390,488 on the reverse strand of chromosome 12 in genome build GRCh38). The biomarker may include ENSG00000224067.2 (Description: A pseudogene similar to a part of HLA-B-associated transcript 2 (BAT2)). The biomarker may include ENSG00000196735.13 (Description: Major histocompatibility complex class II, DQ alpha 1, HLA-DQA1). The biomarker may include ENSG00000287647.1 (Description: Antisense to AK5). The biomarker may include ENSG00000230797.3 (Description: YY2 transcription factor, YY2). The biomarker may include ENSG00000287219.1 (Description: A novel human transcript derived from positions 38,675,876 to 38,677,800 on the forward strand of chromosome 22 in genome build GRCh38). The biomarker may include ENSG00000271543.1 (Description: Ribosomal protein L6 (RPL6) pseudogene) may be included. The biomarker may include ENSG00000223711.1 (Description: AC091633.3 (clone-based (Vega) gene) at positions 195,270,871 to 195,277,400 on the forward strand of chromosome 3 of the human genome build GRCh37). The biomarker may include ENSG00000223711.2 (Description: A novel transcript at positions 195,543,418 to 195,550,581 on the forward strand of chromosome 3 of the genome build GRCh38). The biomarker may include ENSG00000177602.5 (Description: Histone H3-related protein kinase, HASPIN). The biomarker may include ENSG00000144671.11 (Description: Solute carrier family 22 member 14, SLC22A14). The biomarker may include ENSG00000129673.10 (Description: Arylamine N-acetyltransferase, AANAT). The biomarker may include ENSG00000265817.4 (Description: Fibrinogen silencer-binding protein, FSBP). The biomarker may include ENSG00000108924.14 (Description: HLF transcription factor, PAR bZIP family member, HLF). The biomarker may include ENSG00000232125.5 (Description: Distroterin, DYTN). The biomarker may include ENSG00000252800.1 ((Description: A human transcript derived from positions 63,479,272 to 63,479,413 on the forward strand of chromosome 14 of the genome build GRCh38), this RNA biomarker may correspond to small Cajal body-specific RNA 20 (SCARNA20)). The biomarker may include ENSG00000287537.1 (Description: A novel human transcript derived from positions 49,536,677 to 49,538,894 on the reverse strand of chromosome 12 of the genome build GRCh38). The biomarker may include ENSG00000196405.13 (Description: Enah / Vasp-like, EVL). The biomarker may include ENSG00000250893.It may include 1 (Description: A novel human transcript derived from positions 40,426,119 to 40,427,585 on the forward strand of chromosome 4 of genome build GRCh38). The biomarker may include ENSG00000153446.16 (Description: Chromosome 16 open reading frame 89, C16orf89). The biomarker may include ENSG00000284630.1 (Description: A novel human transcript derived from positions 21,657,811 to 21,661,021 on the forward strand of chromosome 22 of genome build GRCh38). The biomarker may include ENSG00000284687.1 (Description: RNA binding protein, fox-1 homolog (C.elegans) 1 (RBFOX1) human pseudogene derived from positions 8,390,270 to 8,390,488 on the reverse strand of chromosome 12 of genome build GRCh38). Any of these biomarkers may be useful, alone or in combination, for assessing a lung nodule (e.g., determining the likelihood that the lung nodule is cancerous). 【0404】 RNA can be used as a biomarker, which includes mRNA. One, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, or twenty of these RNAs may be useful as biomarkers in lung cancer, such as non-small cell lung cancer). Any of the following RNAs may be useful as such (presented as Ensembl reference numbers): ENSG00000155744.10 (description: Hyccin PI4KA lipid kinase complex subunit 2, HYCC2), ENSG00000081052.14 (description: type IV collagen alpha4 chain, COL4A4), ENSG00000173726.11 (description: translocase of outer mitochondrial membrane 20, TOMM20), ENSG00000143995.20 (description: Meis homeobox 1, MEIS1), ENSG00000108528.14 (description: solute carrier family 25 member 11, SLC25A11), ENSG00000177427.13 (description: mitochondrial elongation factor 2, MIEF2), ENSG00000163961.4 (description: Ring finger protein 168, RNF168), ENSG00000049130.16 (description: KIT ligand, KITLG), ENSG00000008405.12 (description: cryptochrome circadian regulator 1, CRY1), ENSG00000135090.14 (description: TAO kinase 3, TAOK3), ENSG00000151778.11 (description: stress-related endoplasmic reticulum protein family member 2, SERP2), ENSG00000172116.23 (description: CD8b molecule, CD8B), ENSG00000144218.21 (description: ALF transcription elongation factor 3, AFF3), ENSG00000131196.18 (description: nuclear factor of activated T cells 1, NFATC1), ENSG00000129351.18 (description: interleukin enhancer binding factor 3, ILF3), ENSG00000105518.14 (description: transmembrane protein 205, TMEM205), ENSG00000182162.11 (description: P2Y receptor family member 8, P2RY8), ENSG00000126368.6 (Description: Nuclear receptor subfamily 1 group D member 1, NR1D1), ENSG00000176358.16 (Description: Tachykinin precursor 4, TAC4), ENSG00000112599.9 (Description: Guanylate cyclase activating factor 1B, GUCA1B), ENSG00000142864.15 (Description: SERPINE1 mRNA binding protein 1, SERBP1), ENSG00000163159.15 (Description: Vacuolar protein sorting 72 homolog, VPS72), ENSG00000165661.17, ENSG00000165661.18 (Description: Quiescin sulfhydryl oxidase 2, QSOX2), ENSG00000007923.17 (Description: DnaJ heat shock protein family (Hsp40) member C11, DNAJC11), ENSG00000054116.12 (Description: Transport protein particle complex subunit 3, TRAPPC3), ENSG00000113811.12 (Description: Selenoprotein K, SELENOK), ENSG00000100644.17 (Description: Hypoxia-inducible factor 1 subunit alpha, HIF1A), ENSG00000133997.12 (Description: Mediator complex subunit 6, MED6), ENSG00000120925.16 (Description: Ring finger protein 170, RNF170), ENSG00000110048.12 (Description: Oxysterol binding protein, OSBP), ENSG00000197863.9 (Description: Zinc finger protein 790, ZNF790), ENSG00000174307.7 (Description: Pleckstrin homology-like domain family A member 3, PHLDA3), or ENSG00000109381.21 (Description: E74-like ETS transcription factor 2, ELF2). The biomarker may include ENSG00000155744.10 (Description: Hyccin PI4KA lipid kinase complex subunit 2, HYCC2). The biomarker may include ENSG00000081052.14 (Description: Type IV coll...

Claims

[Claim 1] A method, (a) To quantify one or more proteins or fragments in a biological fluid sample derived from a subject, wherein the one or more proteins or fragments include complement component C9 (C9), apolipoprotein A4 (APOA4), or both. (b) Obtaining a plurality of proteomics measurements relating to one or more proteins or fragments thereof, wherein the plurality of proteomics measurements are less than 100, and (c) Using a classifier to analyze the multiple proteomics measurements and assign the degree of lung cancer risk in the subject with a sensitivity of at least 75%, wherein the classifier is trained on lung cancer and non-cancerous samples. Methods that include... [Claim 2] The method according to claim 1, wherein the lung cancer includes non-small cell lung cancer (NSCLC). [Claim 3] The method according to claim 1, wherein the subject is at risk of having lung cancer. [Claim 4] The method according to claim 3, wherein the risk of having lung cancer is determined at least in part on the age of the subject, the smoking history of the subject, or a combination thereof. [Claim 5] The method according to claim 1, wherein one or more proteins or fragments thereof are quantified by mass spectrometry, immunoassay, or a combination thereof. [Claim 6] The method according to claim 5, wherein the immunoassay comprises enzyme-linked immunosorbent assay (ELISA). [Claim 7] The method according to claim 1, wherein the plurality of proteomics measurements include the level or amount of one or more proteins or fragments thereof. [Claim 8] The method according to claim 1, further comprising obtaining quantitative or qualitative RNA nucleic acid data of the subject and analyzing it using the classifier, further comprising analyzing the quantitative or qualitative RNA data in addition to the plurality of proteomics measurements using the classifier to assign the degree of lung cancer risk in the subject. [Claim 9] The method according to claim 8, wherein the RNA nucleic acid data includes expression data. [Claim 10] The method according to claim 1, further comprising obtaining levels or amounts of one or more metabolites or fragments thereof to generate a plurality of metabolomics measurements, and analyzing them using the classifier, further comprising analyzing the plurality of metabolic measurements in addition to the plurality of proteomics measurements using the classifier to assign the degree of lung cancer risk in the subject. [Claim 11] The method according to claim 1, wherein the sensitivity is at least 80%. [Claim 12] The method according to claim 1, wherein the sensitivity is at least 85%. [Claim 13] The method according to claim 1, wherein the biological fluid sample is plasma or serum. [Claim 14] The method according to claim 1, wherein the one or more proteins or fragments thereof further comprise fibrinogen-like protein 1 (FGL1) and serum amyloid A (SAA). [Claim 15] The method according to claim 14, wherein the one or more proteins include FGL1. [Claim 16] The method according to claim 14, wherein the one or more proteins include the SAA. [Claim 17] The method according to claim 14, wherein the one or more proteins or fragments thereof further comprise orosomucoid-1 (ORM1), C-reactive protein (CRP), insulin-like growth factor-binding protein 2 (IGFB2), S100 calcium-binding protein A8 (S100A8), and S100 calcium-binding protein A9 (S100A9). [Claim 18] The method according to claim 17, wherein the one or more proteins further comprise two or more selected from orosomucoid-1 (ORM1), C-reactive protein (CRP), insulin-like growth factor-binding protein 2 (IGFB2), S100 calcium-binding protein A8 (S100A8), or S100 calcium-binding protein A9 (S100A9). [Claim 19] The method according to claim 17, wherein the one or more proteins further comprise three or more selected from orosomucoid-1 (ORM1), C-reactive protein (CRP), insulin-like growth factor-binding protein 2 (IGFB2), S100 calcium-binding protein A8 (S100A8), or S100 calcium-binding protein A9 (S100A9). [Claim 20] The method according to claim 17, wherein the one or more proteins further comprise four or more of orosomucoid-1 (ORM1), C-reactive protein (CRP), insulin-like growth factor-binding protein 2 (IGFB2), S100 calcium-binding protein A8 (S100A8), or S100 calcium-binding protein A9 (S100A9).