Combined marker combination and detection method and kit thereof

Abstract

The invention provides a combined marker combination as well as a detection method and a kit thereof. The combined marker combination comprises a plasma protein marker and a blood metabolism marker, the plasma protein marker comprises an adaptor protein PH-containing structural domain, enolase 3, a proline 4-hydroxylase alpha1 chain and ethanol dehydrogenase 1C, and the adaptor protein comprises an adaptor protein PH-containing structural domain, a proline 4-hydroxylase alpha1 chain and a proline 4-hydroxylase alpha2 chain. The blood metabolism marker is prepared from diosgenin, eicosapentaenoic acid, folinic acid, hydroxyprogesterone hexanoate, testosterone, perfluorooctane sulfonic acid, prasterone sulfate, androstane dienone, long pine needle ketone and 1-(22 carbon 4 enoyl) lysophosphatidylcholine. Through cooperative detection of protein-metabolism markers and combined application of a machine learning model, accurate prediction of the age-related disease risk is achieved, the cognitive decline risk can be pre-judged in advance, meanwhile, efficient synchronous detection of multiple markers is achieved by means of a liquid chromatography-tandem mass spectrometry synchronous detection technology, the sample dosage is small, the detection process is standardized, and the detection efficiency is high. The large-scale queue research requirement is met, and multi-center clinical application and popularization are facilitated.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical laboratory and aging biology, in particular to a combined marker combination and a detection method and kit thereof. BACKGROUND

[0002] Currently, the global population aging process is accelerating, and age-related diseases (ARDs) are chronic diseases with a significantly increased incidence with age, including cognitive impairment, cardiovascular events, metabolic syndrome, etc. have become major public health challenges. The existing clinical evaluation system mainly has the following defects: (1) Limitations of traditional evaluation methods: Clinically, subjective scale evaluation (such as Mini-Mental State Examination MMSE) or single biomarker detection (such as C-reactive protein, interleukin-6) is commonly used, which has problems such as early warning lag (irreversible pathological changes often occur at the time of diagnosis), insufficient sensitivity (AUC is usually <0.75), large individual difference interference, etc., and it is difficult to achieve early risk stratification.

[0003] (2) One-sidedness of single omics technology: In recent years, independent proteinomics or metabolomics research has revealed age-related molecular changes (such as 373 plasma proteins related to age reported by Lehallier et al.), but single omics can only capture part of the biological information: protein function changes and metabolite level fluctuations exist in space-time mismatch, and cannot systematically reflect the cascade regulation network of “gene expression-protein function-metabolic phenotype” in the aging process, resulting in a ceiling effect in prediction efficiency.

[0004] (3) Technical bottlenecks in detection process: In the prior art, protein detection (such as ELISA, Olink) and metabolite detection (such as GC-MS, LC-MS) require independent sampling and separate pretreatment, which leads to: ① large sample consumption (usually >500 μL of plasma); ② large batch difference (CV>15%); ③ long detection period (3-5 working days); ④ high cost (single sample >2000 yuan), which seriously restricts the clinical feasibility of large-scale cohort screening.

[0005] (4) Lack of standardization in data analysis: There is a lack of marker combination threshold system and interpretable biological age score model verified by independent cohorts, and the reproducibility of multi-center studies is poor (the reported reproducibility is <40%), which limits the translation process of production, teaching and research. Among them, the biological age score is a quantitative index calculated by combining multiple molecular markers, which reflects the physiological aging degree of individuals, and is different from the actual calendar age.

[0006] In summary, how to realize single sample, low consumption, high throughput protein-metabolite synchronous detection, how to construct a combined marker combination with synergistic prediction value, and how to establish a standardized and repeatable clinical transformation scheme are problems that need to be solved. SUMMARY

[0007] The main purpose of the present application is to provide a combined marker combination and its detection method and kit to at least solve the problems of traditional evaluation and early warning lag, insufficient sensitivity of age-related diseases in the prior art, one-sidedness of single omics detection, limited prediction performance, large sample consumption, large batch difference, long cycle, and high cost of independent detection of proteins and metabolites.

[0008] To achieve the above-mentioned purpose, the first aspect of the present application provides a combined marker combination comprising a plasma protein marker and a blood metabolite marker. The plasma protein marker comprises protein containing PH domain, enolase 3, proline 4-hydroxylase alpha 1 chain, alcohol dehydrogenase 1C, eukaryotic translation initiation factor 5B, phosphomevalonate kinase, STARD3-like protein, carbonic anhydrase 6, cadherin 5, insulin-like growth factor 1, chromatin modification protein 4B, caltrin binding heat shock protein 1, collagen agglutinin 11, serine / threonine kinase 39, mannan-binding lectin-associated serine protease 1, cluster of differentiation 14, cholesterol ester transfer protein, tRNA export protein, signal sequence receptor subunit 1, macrophage scavenger receptor, interferon-induced protein 27, SMAD family member 1, aminoacylase 1, carboxypeptidase N subunit 1, complement component 1q subunit A, cadherin 11, paraoxonase 1, FYN binding protein 1, lysine tRNA synthetase 1, gamma-tubulin complex protein 3, fibril agglutination protein 2, GRB2-related adaptor protein 2, Wiskott-Aldrich syndrome protein-interacting protein family member 1, ubiquitination factor homolog 6, peptidoglycan recognition protein 2, sorting nexin 12, actin-related protein 2 / 3 complex subunit 2, epidermal protein 1. The blood metabolism markers include diosgenin, eicosapentaenoic acid, folinic acid, hydroxyprogesterone caproate, testosterone, perfluorooctane sulfonic acid, prasterone sulfate, androstenedione, long pine needle ketone, 1-(22 carbon 4 ene acyl) lysophosphatidylcholine, 16 carbon lysate platelet activating factor, proline phenylalanine, testosterone sulfate, tuber acid, valyl-prolyl-arginyl-aspartic acid, 5,6-dihydrogen arachidonic acid, 5α-androstan-3β-ol-17-ketoglucuronide, 5α-pregnane-3α, 17-diol-20-ketone-3-sulfate, 3-carboxy-4-methyl-5-propyl-2-furanpropionic acid, 40:6 fatty acid ester fatty acid, 12-hydroxy-9-octadecyne acid, hexadecanedioic acid mono-isopropyl ester, branched C3 perfluoroalkyl sulfonic acid; hexafluorohexyl sulfonic acid, phoma stem ketone, (3α, 5β, 7α)-23-carboxy-7-hydroxy-24-norcholane-3-yl-β-D-glucuronide, 2-methyl-3-hydroxy ethylene pyran-4-ketone, 3-hydroxy-3', 4'-dimethoxy flavone, 4-[(4-fluorophenoxy) methyl]-1-[1-[(3-methylphenyl) methyl] azetidin-3-yl] triazole.

[0009] The second aspect of the present application provides a detection method for detecting the combined marker combination of the first aspect, comprising the following steps: (1) Plasma sample collection and pretreatment: Collect EDTA anticoagulant peripheral venous blood, centrifugal separation of plasma; Take the plasma to join the complex of biotin labeled polyclonal capture antibody and streptavidin magnetic beads for immunocapture, obtain the magnetic bead complex rich in target protein and supernatant containing metabolites; (2) Target protein treatment: The magnetic bead complex obtained in step (1) is washed, eluted, denatured, reduced, alkylated and enzymatically treated to obtain a protein peptide sample; (3) Target metabolite treatment: Add pre-cooled extraction solution and internal standard mixed solution to the supernatant containing metabolites obtained in step (1), vortex, precipitate, centrifuge, concentrate and redissolve to obtain a metabolite sample; (4) Synchronous detection: The protein peptide sample of step (2) and the metabolite sample of step (3) are detected by liquid chromatography-tandem mass spectrometry to obtain the quantitative data of each marker; (5) Risk assessment: The quantitative data obtained in step (4) is input into a machine learning model to calculate a biological age score, and the risk grade of the aging-related disease is determined according to the biological age score.

[0010] Optionally, in step (1), the conditions for immunocapture are 37℃, 1200rpm shaking incubation for 60min; The polyclonal capture antibody comprises anti-APPL1, anti-ENO3, anti-P4HA1, anti-ADH1C, anti-EIF5B, anti-PMVK, anti-STARD3NL, anti-CA6, anti-CDH5, anti-IGF1, anti-CHMP4B, anti-CARHSP1, anti-COLEC11, anti-STK39, anti-MASP1, anti-CD14, anti-CETP, anti-XPOT, anti-SSR1, anti-MARCO, anti-IFI27, anti-SMAD1, anti-ACY1, anti-CPN1, anti-C1QA, anti-CDH11, anti-PON1, anti-FYB1, anti-KARS1, anti-TUBGCP3, anti-FCN2, anti-GRAP2, anti-WIPF1, anti-UBXN6, anti-PGLYRP2, anti-SNX12, anti-ARPC2, anti-EPN1, each in an amount of 2 μg.

[0011] Optionally, in step (1), the amount of the blood plasma is 100 μL; and the centrifugal separation condition of the blood plasma is 4℃, 3000g centrifugal separation for 10 min.

[0012] Optionally, in step (3), the pre-cooled extraction liquid is a solution of methanol, acetonitrile and water mixed at a volume ratio of 2:2:1. The internal standard mixture comprises d4-diosgenin, d5-eicosapentaenoic acid, ¹³C5-folinic acid, d6-hexanoic acid hydroxyprogesterone, d3-testosterone, ¹³C8-perfluorooctanesulfonic acid, d4-prasterone sulfate, d3-androstadienone, d5-longpinoketone, d4-1-(22 carbon 4 enoyl) lysophosphatidylcholine, d3-16 carbon lysate platelet activating factor, d4-prolyl phenylalanine, d3-testosterone sulfate, d4-tuberonic acid, ¹³C4-valyl-prolyl-arginyl-aspartic acid, d5-5,6-dihydroarachidonic acid, ¹³C3-5α-androstan-3β-ol-17-ketoglucuronide, d4-5α-pregnane-3α,17-diol-20-ketone-3-sulfate, d3-3-carboxy-4-methyl-5-propyl-2-furanpropionic acid, d6-40:6 fatty acid ester fatty acid, d4-12-hydroxy-9-octadecyne acid, d5-hexadecanedioic acid monoisopropyl ester, ¹³C4-branched C3 perfluoroalkyl sulfonic acid; hexafluorohexyl sulfonic acid, d4-stem point ketone, ¹³C5-(3α,5β,7α)-23-carboxy-7-hydroxy-24-norcholane-3-yl-β-D-glucuronide, d3-2-methyl-3-hydroxyethylpyran-4-ketone, d4-3-hydroxy-3',4'-dimethoxyflavone, d5-4-[(4-fluorophenoxy)methyl]-1-[1-[(3-methylphenyl)methyl]azetidin-3-yl]triazole, each internal standard at a final concentration of 100 ng / mL.

[0013] Optionally, in step (5), the machine learning model is a random forest model or a LASSO regression model.

[0014] In a third aspect, the present application provides a kit for predicting the risk of an aging-related disease, comprising corresponding detection reagents of the combined marker combination of the first aspect of the present application, wherein the detection reagents comprise biotin-labeled polyclonal capture antibodies, streptavidin magnetic beads, washing buffer, eluent, pre-cooled extraction solution, internal standard mixture, reconstitution solution, calibrators and quality control samples.

[0015] In a fourth aspect, the present application provides use of the combined marker combination of the first aspect of the present application in the preparation of a product for predicting the risk of an aging-related disease.

[0016] The application discloses a combined marker combination and a detection method and kit thereof, and the combined marker combination comprises a plasma protein marker and a blood metabolism marker; the plasma protein marker comprises protein containing a PH domain, enolase 3, proline 4-hydroxylase alpha 1 chain, alcohol dehydrogenase 1C, eukaryotic translation initiation factor 5B, phosphomevalonate kinase, STARD3-like protein, carbonic anhydrase 6, calcium adhesion protein 5, insulin-like growth factor 1, chromatin modification protein 4B, calmodulin-binding heat shock protein 1, collagen agglutinin 11, serine / threonine kinase 39, mannan-binding lectin-related serine protease 1, cluster of differentiation 14, cholesterol ester transfer protein, tRNA export protein, signal sequence receptor subunit 1, macrophage scavenger receptor, interferon-induced protein 27, SMAD family member 1, aminoacylase 1, carboxypeptidase N subunit 1, complement component 1q subunit A, calcium adhesion protein 11, paraoxonase 1, FYN-binding protein 1, lysine tRNA synthetase 1, gamma-tubulin complex protein 3, fibrous agglutination protein 2, GRB2-related adaptor protein 2, Wiskott-Aldrich syndrome protein-interacting protein family member 1, ubiquitinylating factor homolog 6, peptidoglycan recognition protein 2, sorting nexin 12, actin-related protein 2 / 3 complex subunit 2, epidermal protein 1; and the blood metabolism marker comprises diosgenin, eicosapentaenoic acid, folinic acid, hydroxyprogesterone caproate, testosterone, perfluorooctane sulfonate, prasterone sulfate, androstadienone, long pine needle ketone, 1-(22 carbon 4 enoyl) lysophosphatidylcholine, 16 carbon lysophosphatidylcholine, proline phenylalanine, testosterone sulfate, tuber acid, valyl-prolyl-arginyl-aspartic acid, 5,6-dihydroarachidonic acid, 5alpha-androstan-3beta-ol-17-ketoglucuronide, 5alpha-pregnane-3alpha, 17-diol-20-ketone-3-sulfate, 3-carboxy-4-methyl-5-propyl-2-furanpropionic acid, 40:6 fatty acid ester fatty acid, 12-hydroxy-9-octadecyne acid, hexadecanedioic acid monoisopropyl ester, branched C3 perfluoroalkyl sulfonic acid, hexafluorohexyl sulfonic acid, phomopsidone, (3alpha, 5beta, 7alpha)-23-carboxy-7-hydroxy-24-norcholane-3-yl-beta-D-glucuronide, 2-methyl-3-hydroxyethylidene pyran-4-ketone, 3-hydroxy-3', 4'-dimethoxy flavone, 4-[(4-fluorophenoxy) methyl]-1-[1-[(3-methylphenyl) methyl] azetidin-3-yl] triazole. Through the combined detection of protein-metabolism markers and the joint application of a machine learning model, the risk of aging-related diseases can be accurately predicted, the risk of cognitive decline can be predicted in advance, and meanwhile, the synchronous detection of multiple markers is realized by relying on liquid chromatography-tandem mass spectrometry, sample consumption is small, and the detection process is standardized, so that the demand of large-scale cohort study is met, and the application in multi-center clinical promotion is facilitated. BRIEF DESCRIPTION OF DRAWINGS

[0017] The accompanying drawings, which form a part of this specification, are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification. The embodiments of the application, together with its Figure 1 is a flow chart of an optional detection method according to an embodiment of the application. DETAILED DESCRIPTION

[0018] It should be noted that the embodiments and features of the present application can be combined with each other, if not in conflict. The present application will be described in further detail with reference to the drawings and embodiments.

[0019] The present application provides a combined marker combination, comprising a plasma protein marker and a blood metabolite marker; The plasma protein marker comprises adaptor protein containing PH domain (APPL1), enolase 3 (ENO3), proline 4-hydroxylase alpha 1 chain (P4HA1), alcohol dehydrogenase 1C (ADH1C), eukaryotic translation initiation factor 5B (EIF5B), phosphomevalonate kinase (PMVK), STARD3-like protein (STARD3NL), carbonic anhydrase 6 (CA6), cadherin 5 (CDH5), insulin-like growth factor 1 (IGF1), chromatin-modifying protein 4B (CHMP4B), caltractin-binding heat shock protein 1 (CARHSP1), collagen, lectin 11 (COLEC11), serine / threonine kinase 39 (STK39), mannan-binding lectin-associated serine protease 1 (MASP1), cluster of differentiation 14 (CD14), cholesteryl ester transfer protein (CETP), tRNA export protein (XPOT), signal sequence receptor subunit 1 (SSR1), macrophage scavenger receptor (MARCO), interferon-inducible protein 27 (IFI27), SMAD family member 1 (SMAD1), aminoacylase 1 (ACY1), carboxypeptidase N subunit 1 (CPN1), complement component 1q subunit A (C1QA), cadherin 11 (CDH11), paraoxonase 1 (PON1), FYN-binding protein 1 (FYB1), lysine tRNA synthetase 1 (KARS1), gamma-tubulin complex protein 3 (TUBGCP3), fibrillar collagen 2 (FCN2), GRB2-associated adaptor protein 2 (GRAP2), Wiskott-Aldrich syndrome protein-interacting protein family member 1 (WIPF1), ubiquitin- ligase homolog 6 (UBXN6), peptidoglycan recognition protein 2 (PGLYRP2), sorting nexin 12 (SNX12), actin-related protein 2 / 3 complex subunit 2 (ARPC2), epithelial protein 1 (EPN1); The blood metabolic markers include diosgenin, eicosapentaenoic acid, folinic acid, hydroxyprogesterone caproate, testosterone, perfluorooctanesulfonic acid, prasterone sulfate, androstadienone, longicamphenylone, 1-(22 carbon 4 ene acyl) lysophosphatidylcholine (LPC22:4-SN1), 16 carbon lysophosphatidylcholine (Lyso-PAFC-16), proline phenylalanine (ProPhe), testosterone sulfate, tuberonic acid, valine-proline-arginine-aspartic acid (ValProArgAsp), 5,6-dehydroarachidonic acid, 5α-androstan-3β-ol-17-one glucosiduronate, 5α-pregnan-3α, 17-diol-20-one 3-sulfate, 3-carboxy-4-methyl-5-propyl-2-furanpropionic acid (CMPF), 40:6 fatty acid ester fatty acid (FAHFA40:6), 12-hydroxy-9-octadecynoic acid, hexadecanedioic acid mono isopropyl ester (MEDICA16), branched C3 perfluoroalkyl sulfonic acid; hexafluorohexyl sulfonic acid (PFSA-perfluoroalkyl_branched_C3;C6HF13O3S), phomalone, (3a, 5b, 7a)-23-carboxy-7-hydroxy-24-norcholan-3-yl- b-D-glucopyranosiduronic acid, 2-methyl-3-hydroxyethylenepyran-4-one, 3-hydroxy-3', 4'-dimethoxyflavone, 4-[(4-fluorophenoxy)methyl]-1-[1-[(3-methylphenyl)methyl]azetidin-3-yl]triazole.

[0020] The combined marker combination is screened based on large-scale cohort proteomics-metabolomics correlation analysis, and ensures that the markers have a synergistic predictive value in aging-related pathways (oxidative stress, inflammation, energy metabolism).

[0021] As shown in Figure 1 The present application provides a detection method for detecting the combined marker combination of the present application, comprising the following steps: (1) Plasma sample collection and pretreatment: Collect EDTA anticoagulant peripheral venous blood, centrifugal separation of plasma; take the plasma to join the complex of biotin labeled polyclonal capture antibody and streptavidin magnetic beads to carry out immunocapture, obtain the magnetic bead complex rich in target protein and supernatant containing metabolites; wherein, the immunocapture is a technology for selectively enriching target proteins from complex plasma samples by using antibody-antigen specific binding principle.

[0022] PRM / MRM: parallel reaction monitoring / multiple reaction monitoring, mass spectrometry targeted quantitative technology, with high sensitivity and specificity.

[0023] (2) Target protein processing: the magnetic bead complex obtained in step (1) is washed, eluted, denatured, reduced, alkylated and enzymatically treated to obtain a protein peptide sample; (3) Target metabolite processing: adding pre-cooled extraction solution and internal standard mixed solution to the supernatant containing metabolites obtained in step (1), vortexing, standing precipitation, centrifugation, concentration and resuspension treatment to obtain a metabolite sample; (4) Synchronous detection: the protein peptide sample of step (2) and the metabolite sample of step (3) are detected by liquid chromatography-tandem mass spectrometry, respectively, to obtain quantitative data of each marker; (5) Risk assessment: input the quantitative data obtained in step (4) into a machine learning model to calculate a biological age score, and determine the risk grade of the aging-related disease according to the biological age score.

[0024] In step (1), the conditions for the immune capture are 37℃, 1200rpm shaking incubation for 60min; The polyclonal capture antibody comprises anti-APPL1, anti-ENO3, anti-P4HA1, anti-ADH1C, anti-EIF5B, anti-PMVK, anti-STARD3NL, anti-CA6, anti-CDH5, anti-IGF1, anti-CHMP4B, anti-CARHSP1, anti-COLEC11, anti-STK39, anti-MASP1, anti-CD14, anti-CETP, anti-XPOT, anti-SSR1, anti-MARCO, anti-IFI27, anti-SMAD1, anti-ACY1, anti-CPN1, anti-C1QA, anti-CDH11, anti-PON1, anti-FYB1, anti-KARS1, anti-TUBGCP3, anti-FCN2, anti-GRAP2, anti-WIPF1, anti-UBXN6, anti-PGLYRP2, anti-SNX12, anti-ARPC2, anti-EPN1, and the amount of each antibody is 2ug.

[0025] In step (1), the amount of the plasma is 100ul; and the conditions for centrifugal separation of the plasma are 4℃, 3000g centrifugation for 10min.

[0026] In step (3), the pre-extraction solution is a solution of methanol, acetonitrile and water mixed at a volume ratio of 2:2:1; the internal standard mixture contains d4-diosgenin, d5-eicosapentaenoic acid, 13C5-folinic acid, d6-hexafluoroisopropyl alcohol, d3-testosterone, 13C8-perfluorooctanesulfonic acid, d4-prasteride sulfate, d3-androstadienone, d5-longpinenone, d4-1-(22 carbon 4 ene acyl) lysophosphatidylcholine, d3-16 carbon lysophosphatidylcholine, d4-prolyl phenylalanine, d3-testosterone sulfate, d4-tuberonic acid, 13C4-valyl-prolyl-arginyl-aspartic acid, d5-5,6-dehydroarachidonic acid, 13C3-5α-androstan-3β-ol-17-ketoglucuronide, d4-5α-pregnane-3α, 17-diol-20-ketone-3-sulfate, d3-3-carboxy-4-methyl-5-propyl-2-furanpropionic acid, d6-40:6 fatty acid ester fatty acid, d4-12-hydroxy-9-octadecyne acid, d5-hexadecanedioic acid monoisopropyl ester, 13C4-branched C3 perfluoroalkyl sulfonic acid; hexafluorohexyl sulfonic acid, d4-stem point mold ketone, 13C5-(3α, 5β, 7α)-23-carboxy-7-hydroxy-24-norcholane-3-yl-β-D-glucuronide, d3-2-methyl-3-hydroxy ethylene pyran-4-ketone, d4-3-hydroxy-3', 4'-dimethoxy flavone, d5-4-[(4-fluorophenoxy) methyl]-1-[1-[(3-methylphenyl) methyl] azetidin-3-yl] triazole, and the final concentration of each internal standard is 100 ng / mL.

[0027] In step (5), the machine learning model is a random forest model or a LASSO regression model.

[0028] Specifically, the simultaneous detection of proteins and metabolites is achieved by using the immunocapture-LC-MS / MS combined technology, including the following steps: a. Blood sample collection and pretreatment Anticoagulant peripheral venous blood is collected, and plasma is separated by centrifugation (4℃ 3000g, 10min). Take 100μL of plasma for hierarchical processing: Protein component: Take 100µL of blood sample after high-speed centrifugation into the pre-washed magnetic nanomaterial, place it in a constant temperature homogenizer at 1200rpm, 37℃ incubate for 1 hour. After incubation, wash the magnetic beads with washing buffer for 3 times. Add 150μL of enzyme digestion buffer to the magnetic beads, add trypsin with a final concentration of 10ng / μL, and incubate at 37℃ overnight for enzyme digestion; add dithiothreitol (DTT) to make its final concentration 5mM, and reduce at 56℃ for 30min. Then add iodoacetamide (IAM) to make its final concentration 11mM, and incubate at room temperature for 15min in the dark. Desalt according to the C18ZipTips instruction, and vacuum freeze-dry for liquid chromatography-mass spectrometry analysis; Protein component enrichment: Magnetic bead activation: Take 50 μL of streptavidin magnetic beads (10 mg / mL), wash with 200 μL of PBS (pH 7.4) for 2 times; Antibody coupling: Add biotin-labeled polyclonal capture antibody mixture (anti-APPL1, anti- ENO3, anti-P4HA1, anti-ADH1C, anti-EIF5B, anti-PMVK, anti-STARD3NL, anti-CA6, anti- CDH5, anti-IGF1, anti-CHMP4B, anti-CARHSP1, anti-COLEC11, anti-STK39, anti-MASP1, anti- CD14, anti-CETP, anti-XPOT, anti-SSR1, anti-MARCO, anti-IFI27, anti-SMAD1, anti-ACY1, anti- CPN1, anti-C1QA, anti-CDH11, anti-PON1, anti-FYB1, anti-KARS1, anti-TUBGCP3, anti-FCN2, anti- GRAP2, anti-WIPF1, anti-UBXN6, anti-PGLYRP2, anti-SNX12, anti-ARPC2, anti-EPN1, 2 μg of each), incubate at room temperature for 30 min, and wash with PBS; Immune capture: Add the magnetic bead-antibody complex to the plasma sample, incubate at 37°C with 1200 rpm shaking for 60 min, and discard the supernatant after magnetic field separation; Washing: Wash with 200 μL of washing buffer (50 mM Tris-HCl, 150 mM NaCl, 0.1% Tween-20, pH 7.5) for 3 times; Elution: Add 50 μL of 0.1 M glycine-HCl (pH 2.5) for elution, and immediately neutralize with 1 M Tris.

[0029] Metabolite extraction (synchronized with immune capture, washing, and elution): The supernatant (containing metabolites) after immune capture was transferred to a new tube, 400 μL of pre-cooled extraction solution (MeOH:ACN:H2O=2:2:1, v / v / v) containing internal standard mixture (d4-Diosgenin, d5-Eicosapentaenoic acid, 13C5-Folinic acid, d6-Hydroxyprogesterone caproate, d3-Testosterone, 13C8-Perfluorooctanesulfonic acid, d4-Prasterone sulfate, d3-Androstadienone, d5-Longicamphenylone, d4-LPC22:4-SN1, d3-Lyso-PAFC-16, d4-ProPhe, d3-Testosterone sulfate, d4-Tuberonic acid, 13C4-ValProArgAsp, d5-5,6-dehydroArachidonic Acid, 13C3-5α-Androstan-3β-ol-17-one glucosiduronate, d4-5α-Pregnan-3α,17-diol-20-one 3-sulfate, d3-CMPF, d6-FAHFA40:6, d4-12-hydroxy-9-octadecynoic acid, d5-MEDICA16, 13C4-PFSA-perfluoroalkyl_branched_C3;C6HF1303S, d4-Phomalone, 13C5-(3α,5β,7α)-23-Carboxy-7-hydroxy-24-norcholan-3-yl-β-D-Glucopyranosiduronic acid, d3-2-METHYL-3-HYDROXYETHYLENEPYRAN-4-ONE, d4-3-Hydroxy-3',4'-Dimethoxyflavone, d5-4-[(4-fluorophenoxy)methyl]-1-[1-[(3-methylphenyl)methyl]azetidin-3-yl]triazole), final concentration 100 ng / mL) was added; The mixture obtained in the previous step was vortexed for 30 s and precipitated at -20°C for 1 h; The mixture after standing was centrifuged at 18000 x g at 4°C for 15 min, and the supernatant after centrifugation was transferred to a concentration tube and concentrated to dryness by vacuum centrifugation; To the concentrated to dry metabolite residue, 100 μL of reconstitution solution (ACN:H2O=1:1, containing 0.1% formic acid) was added for reconstitution, and centrifuged at 18000xg for 10 min at 4°C. The supernatant after reconstitution and centrifugation was taken for mass spectrometry detection.

[0030] b. Detection condition configuration Protein detection: After the peptide segment was dissolved with mobile phase A, it was separated using an EASY-nLC1200 ultra-high performance liquid system. Mobile phase A was a solution of 0.1% formic acid and 2% acetonitrile in water; mobile phase B was a solution of 0.1% formic acid and 90% acetonitrile in water. The liquid phase gradient was set as follows: 0-16 min, 6%-20% B; 16-24 min, 20%-32% B; 24-27 min, 32%-80% B; 27-30 min, 80% B, and the flow rate was maintained at 500 nl / min.

[0031] After the peptide segment was separated by the ultra-high performance liquid system, it was injected into the NSI ion source for ionization, and then entered the Orbitrap Exploris480 mass spectrometer for analysis. The ion source voltage was set to 2100V, and both the peptide segment parent ion and its secondary fragments were detected and analyzed using high-resolution Orbitrap. The primary mass spectrometry scan range was set to 350-1050 m / z, and the scan resolution was set to 30000; the secondary mass spectrometry scan range was fixed to 200 m / z, and the secondary scan resolution was set to 45000. The data acquisition mode used the data-independent scanning (DIA) program, i.e. after the primary scan, the peptide segment ions in multiple consecutive m / z windows entered the HCD collision cell using 25%, 30%, and 35% fragmentation energy for fragmentation, and the secondary mass spectrometry analysis was performed in turn. In order to improve the effective utilization rate of mass spectrometry, the automatic gain control (AGC) was set to 3E6, and the maximum injection time was set to Auto.

[0032] Metabolite detection: Peptides were dissolved in mobile phase A and separated using a NanoElute ultra-high performance liquid system. Mobile phase A was a solution of 0.1% formic acid and 2% acetonitrile in water; mobile phase B was a solution of 0.1% formic acid in acetonitrile-water. The liquid phase gradient was set as follows: 0-18 min, 6%-22% B; 18-22 min, 22%-32% B; 22-26 min, 32%-80% B; 26-30 min, 80% B, with a flow rate of 450 nl / min. After separation by the ultra-high performance liquid system, the peptides were injected into the capillary ion source for ionization and then into the timsTOF Pro2 mass spectrometer for analysis. The ion source voltage was set to 1.7 kV, and both the peptide parent ions and their secondary fragments were detected and analyzed using TOF. The secondary mass spectrum scan range was set to 100-1700. The data acquisition mode used the parallel reaction monitoring-parallel cumulative serial fragmentation (prm-PASEF) mode, and the secondary spectrum of the parent ion charge number in the range of 0-5 was collected.

[0033] c. Standard curve establishment and quality control Data processing and quantitative analysis Protein quantification: The PRM chromatographic peak area was extracted using Skyline software, and the isotopically labeled peptides (AQUA peptides) were used as internal standards for absolute quantification; Standard curve: The characteristic peptides of the target protein were synthesized and prepared into a series of concentrations of 5-5000 fmol / μL, with R²≥0.99; Quality control: One QC sample (mixed plasma) was inserted every 10 samples, and a CV<10% was the acceptance standard.

[0034] Metabolite quantification: Peak extraction and alignment were performed using MS-DIAL software, and isotopic internal standards were used for correction; Standard curve: Each metabolite standard was prepared into a series of concentrations of 1-1000 ng / mL, with R 2 ≥0.995; Lower limit of quantification (LLOQ): S / N>10, accuracy 85-115%.

[0035] d. Results integration analysis The protein marker expression level and the metabolite marker concentration were input into a machine learning model (such as random forest or LASSO regression) to calculate the individual "biological age score" and disease risk probability.

[0036] The application further provides a kit for predicting the risk of an aging-related disease, comprising corresponding detection reagents of the combined marker combination of the application, wherein the detection reagents comprise biotin-labeled polyclonal capture antibodies, streptavidin magnetic beads, a washing buffer, an eluent, a pre-cooled extraction solution, an internal standard mixture, a reconstitution solution, a calibrator and a quality control.

[0037] The application further provides an application of the combined marker combination in the preparation of a product for predicting the risk of an aging-related disease.

[0038] The application further provides an application of the combined marker combination in the preparation of a product for predicting the risk of an aging-related disease, wherein the product for predicting the risk of an aging-related disease comprises a protein marker detection module and a metabolic marker detection module.

[0039] The application is further illustrated by the following examples.

[0040] Example 1: Large-scale cohort construction and marker screening Healthy control group: n=150, aged 20-50, one age layer every 10 years, 50 cases in each age layer; Clinical data were collected synchronously: Cognitive function: MoCA scale, digit symbol substitution test (DSST); Vascular function: carotid IMT, etc. Physiological indicators: vital capacity, oxidative stress, etc. 5 mL of fasting EDTA anticoagulant peripheral venous blood was collected, centrifuged at 3000 g at 4℃ for 10 minutes, and the plasma was separated and stored at -80℃.

[0041] Proteomics screening Data-independent acquisition (DIA) mode was adopted, and timsTOF Pro2 mass spectrometer was used for detection. After age correction, LASSO regression screening was performed, and the importance score was evaluated by random forest algorithm, and finally APPL1, ENO3, P4HA1, ADH1C, EIF5B, PMVK, STARD3NL, CA6, CDH5, IGF1, CHMP4B, CARHSP1, COLEC11, STK39, MASP1, CD14, CETP, XPOT, SSR1, MARCO, IFI27, SMAD1, ACY1, CPN1, C1QA, CDH11, PON1, FYB1, KARS1, TUBGCP3, FCN2, GRAP2, WIPF1, UBXN6, PGLYRP2, SNX12, ARPC2, EPN1 were determined as core markers.

[0042] Metabolomics screening Using HILIC-TOF MS non-targeted analysis, positive and negative ion mode, Mann-Whitney U test and Benjamini-Hochberg correction, Diosgenin, Eicosapentaenoic acid, Folinic acid, Hydroxyprogesterone caproate, Testosterone, Perfluorooctanesulfonic acid, prasterone sulfate, Androstadienone, Longicamphenylone, LPC 22:4-SN1, Lyso-PAFC-16, ProPhe, Testosterone sulfate, Tuberonic acid, ValProArgAsp, 5,6-dehydro Arachidonic Acid, 5a-Androstan-3b-ol-17-one glucosiduronate, 5a-Pregnan-3a,17-diol-20-one 3-sulfate, CMPF, FAHFA 40:6, 12-hydroxy-9-octadecynoic acid, MEDICA 16, PFSA-perfluoroalkyl_branched_C3;C6HF1303S, Phomalone, (3a,5b,7a)-23-Carboxy-7-hydroxy-24-norcholan-3-yl-b-D-Glucopyranosiduronic acid, 2-METHYL-3-HYDROXYETHYLENEPYRAN-4-ONE, 3-Hydroxy-3',4'-Dimethoxyflavone, 4-[(4-fluorophenoxy)methyl]-1-[1-[(3-methylphenyl)methyl]azetidin-3-yl]triazole were the core markers.

[0043] Example 2: Immuno-capture-LC-MS / MS method validation Take 100 μL of plasma, add biotin-labeled capture antibody (anti-APPL1, anti- ENO3, anti-P4HA1, anti-ADH1C, anti-EIF5B, anti-PMVK, anti-STARD3NL, anti-CA6, anti- CDH5, anti-IGF1, anti-CHMP4B, anti-CARHSP1, anti-COLEC11, anti-STK39, anti-MASP1, anti- CD14, anti-CETP, anti-XPOT, anti-SSR1, anti-MARCO, anti-IFI27, anti-SMAD1, anti-ACY1, anti- CPN1, anti-C1QA, anti-CDH11, anti-PON1, anti-FYB1, anti-KARS1, anti-TUBGCP3, anti-FCN2, anti- GRAP2, anti-WIPF1, anti-UBXN6, anti-PGLYRP2, anti-SNX12, anti-ARPC2, anti-EPN1, 2 μg of each), and combine with streptavidin magnetic beads; after washing with PBS, denature in 50 mM ammonium bicarbonate (containing 8 M urea), DTT reduction, IA alkylation; after trypsin digestion, nLC-HRMS detection.

[0044] Method performance: LOQ is 5 ng / mL, linear range is 5-5000 ng / mL, recovery rate is 88-105%, intra-batch CV <6%, inter-batch CV <9%.

[0045] Example 3: Construction of cognitive decline risk prediction model APPL1, EN03, P4HA1, ADH1C, EIF5B, PMVK, STARD3NL, CA6, CDH5, IGF1, CHMP4B, CARHSP1, COLEC11, STK39, MASPl, CD14, CETP, XPOT, SSR1, MARCO, IFI27, SMAD1, ACY1, CPN1, C1QA, CDH11, PON1, FYB1, KARS1, TUBGCP3, FCN2, GRAP2, WIPF1, UBXN6, PGLYRP2, SNX12, ARPC2, EPN1, Diosgenin, Eicosapentaenoic acid, Folinic acid, Hydroxyprogesterone caproate, Testosterone, Perfluorooctanesulfonic acid, prasterone sulfate, Androstadienone, Longicamphenylone, LPC 22:4-SN1, Lyso-PAFC-16, ProPhe, Testosterone sulfate, Tuberonic acid, ValProArgAsp, 5,6-dehydroArachidonic Acid, 5a-Androstan-3b-ol-17-one glucosiduronate, 5a-Pregnan-3a,17-diol-20-one 3-sulfate, CMPF, FAHFA 40:6, 12-hydroxy-9-octadecynoic acid, MEDICA 16, PFSA-perfluoroalkyl_branched_C3;C6HF1303S, Phomalone, (3a,5b,7a)-23-Carboxy-7-hydroxy-24-norcholan-3-yl-P-D-Glucopyranosiduronic acid, 2-METHYL-3-HYDROXYETHYLENEPYRAN-4-ONE, 3-Hydroxy-3',4'-Dimethoxyflavone, 4-[(4-fluorophenoxy)methyl]-1-[1-[(3-methylphenyl)methyl]azetidin-3-yl]triazole.

[0046] 3.1 Cohort setup and sample information Verification cohort: 150 healthy subjects were recruited, aged 20-50 years old, without history of nervous system disease, mental illness or major chronic disease; Brain age assessment: All subjects received structural MRI scan at baseline, and a deep learning brain age prediction model trained in large population was used to calculate individual brain age (Brain-Predicted Age); Grouping basis: According to actual age, the subjects were divided into three groups: 20-30 years old, 31-40 years old, and 41-50 years old; According to the difference between brain age and actual age (BrainAGE=brain age actual age), the subjects were divided into: normal aging group (BrainAGE<3 years old), accelerated aging group (BrainAGE≥3 years old); Follow-up design: All subjects were followed up for 3 years, and standardized cognitive assessment was performed every year (including MoCA, AVLT, Stroop, etc.), and the changes in cognitive trajectory were recorded; Sample processing: Fasting EDTA anticoagulation blood was collected, plasma was extracted according to the above method, and the following 66 markers were detected simultaneously: Protein markers: APPL1, ENO3, P4HA1, ADH1C, EIF5B, PMVK, STARD3NL, CA6, CDH5, IGF1, CHMP4B, CARHSP1, COLEC11, STK39, MASP1, CD14, CETP, XPOT, SSR1, MARCO, IFI27, SMAD1, ACY1, CPN1, C1QA, CDH11, PON1, FYB1, KARS1, TUBGCP3, FCN2, GRAP2, WIPF1, UBXN6, PGLYRP2, SNX12, ARPC2, EPN1; Metabolic markers: Diosgenin, Eicosapentaenoic acid, Folinic acid, Hydroxyprogesterone caproate, Testosterone, Perfluorooctanesulfonic acid, prasterone sulfate, Androstadienone, Longicamphenylone, LPC 22:4-SN1, Lyso-PAFC-16, ProPhe, Testosterone sulfate, Tuberonic acid, ValProArgAsp, 5,6-dehydroArachidonic Acid, 5a-Androstan-3b-ol-17-one glucosiduronate, 5a-Pregnan-3a,17-diol-20-one 3-sulfate, CMPF, FAHFA 40:6, 12-hydroxy-9-octadecynoic acid, MEDICA 16, PFSA-perfluoroalkyl_branched_C3;C6HF1303S, Phomalone, (3a,5b,7a)-23-Carboxy-7-hydroxy-24-norcholan-3-yl-beta-D-Glucopyranosiduronic acid, 2-METHYL-3-HYDROXYETHYLENEPYRAN-4-ONE, 3-Hydroxy-3',4'-Dimethoxyflavone, 4-[(4-fluorophenoxy)methyl]-1-[1-[(3-methylphenyl)methyl]azetidin-3-yl]triazole.

[0047] Note: The above 66 markers were found to be significantly abnormal in the accelerated brain aging (BrainAGE>=3 years) population, suggesting their association with neurodegenerative pathways.

[0048] 3.2 Data preprocessing Standardize the raw mass spectrometry / proteomic data (use QC samples for batch correction); Z-score standardization was performed on the log-transformed marker concentration / expression; Calculate the brain age difference (BrainAGE=brain age actual age) for each individual, and use it as a binary classification label: Low-risk group: BrainAGE<3 years; High-risk group: BrainAGE≥3 years.

[0049] 3.3 Model construction procedure Feature selection: LASSO regression was used to reduce the dimensionality of the 10 candidate markers, retaining non-zero coefficient variables that had significant discriminant power for BrainAGE≥3 years; Algorithm training: a binary classification model (low risk vs. high risk) was constructed using a Random Forest classifier; Maximum number of features max_features=(total number of features); 5-fold cross-validation was used to optimize hyperparameters; Output indicators: Biological Age Score (BAS): the continuous risk probability value (0-1) output by the model was linearly mapped to 30-100 points after logit transformation; Risk level determination: BAS<60: low risk (brain age close to actual age); 60≤BAS<75: medium risk (possible mild acceleration of brain aging); BAS≥75: high risk (brain age advanced by≥3 years, suggesting an increased potential risk of cognitive decline).

[0050] 3.4 Model performance verification Discriminatory ability: in the independent verification cohort (n=150), the model predicted BrainAGE≥3 years with an AUC≥0.80, significantly better than single protein models (AUC≥0.60) or single metabolic models (AUC≥0.60); Predictive advance: individuals with high BAS (≥75) had a significantly faster rate of cognitive decline over a 3-year follow-up period than the low-risk group (p<0.01), confirming that brain age difference can be used as a prospective risk indicator.

[0051] 3.5 Clinical application method Integrate the model into the IVD kit analysis software; After the user inputs the 66 marker detection results and the actual age, the system automatically outputs the BAS and risk level; For individuals with BAS≥75, it is suggested that "brain age is significantly advanced", and it is recommended to: Further perform neuroimaging examination (such as MRI brain structure scan); Conduct detailed neuropsychological assessment; Start lifestyle intervention (such as sleep optimization, aerobic exercise, Mediterranean diet) or include cognitive protection clinical trials. Example 4: IVD kit assembly Kit composition: Protein capture module: lyophilized capture antibody (2x96 human portions), magnetic beads, eluent; Metabolite detection module: isotope internal standard mixture, precipitant, reconstitution solution; Calibrators and quality controls: 5-level calibrators, 3-level quality control plasma; Instructions: including standard operating procedures, threshold reference range.

[0052] Stability test: 2-8℃ storage for 12 months, activity of each component decreases by <10%.

[0053] From the above, the present application realizes significant improvement in prediction performance by constructing a specific plasma protein-metabolite marker combined prediction model and combining random forest or LASSO regression machine learning algorithm. The combined model AUC reaches more than 0.8, which is more than 10 percentage points higher than a single marker. The prediction of cognitive decline can be advanced by 3-5 years. By synergistically analyzing the expression changes of protein markers and metabolite markers in the combined marker combination, the biological mechanism is mutually verified, and individual difference interference is effectively reduced. By using liquid chromatography-tandem mass spectrometry synchronous detection technology, high sample throughput is achieved, and ≥5 proteins + ≥8 metabolites can be quantified simultaneously in a single run, with a detection cycle of ≤90 minutes / sample. By optimizing the plasma sample pretreatment process (including EDTA anticoagulation peripheral venous blood collection, centrifugal separation of plasma, and immune capture enrichment of target proteins), only 100 μL of single sample plasma is needed to complete the detection, realizing sample dosage economy, suitable for large-scale cohort study and clinical routine detection. By establishing a standard operating procedure (SOP) covering the whole process of sample collection, pretreatment, detection and data analysis, the inter-laboratory quality evaluation coefficient of variation CV is <12%, which guarantees the standardization and repeatability of the technology, and facilitates the popularization and application of multiple centers.

[0054] It should be noted that the protein-metabolism marker collaborative prediction model of the present application can be achieved by various similar ways, for example, the protein enrichment technology can be replaced by aptamer capture (Aptamer), size exclusion chromatography (SEC) or microfluidic chip immune capture, the detection platform can adopt high-resolution mass spectrometry platform (Q-TOF, Orbitrap Exploris240) or ultra-sensitive immune analysis (Simoa, ECLIA) combined with mass spectrometry, and other aging-related markers such as p16INK4a, IL-6, TMAO can be added on the basis of the existing 66 markers to carry out combination optimization; the application field of the present application is wide, which can be applied not only to clinical departments such as geriatrics, neurology, cardiovascular medicine, physical examination center, but also to biological age assessment of high-end physical examination, anti-aging clinic and other health management scenes, and is suitable for dynamic monitoring of health status of the elderly in pension institutions, research on aging mechanism and longevity cohort, and scientific research services, and health insurance underwriting and risk actuarial work in the insurance industry.

[0055] The above is only the preferred embodiment of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A combination of joint markers, characterized in that, Including plasma protein markers and blood metabolic markers; The plasma protein biomarkers include proteins containing a pH domain, enolase 3, proline 4-hydroxylase α1 chain, alcohol dehydrogenase 1C, eukaryotic translation initiation factor 5B, mevalonate phosphate kinase, STARD 3-like protein, carbonic anhydrase 6, cadherin 5, insulin-like growth factor 1, chromatin modification protein 4B, calmodulin-binding heat shock protein 1, collagen lectin 11, serine / threonine kinase 39, mannan-binding lectin-associated serine protease 1, differentiation cluster 14, cholesterol ester transfer protein, tRNA export protein, signal sequence receptor subunit 1, and macrophage scavenging. The following proteins are included:

1. Receptor, 2. Interferon-inducible protein 27, SMAD family member 1, aminoacylase 1, carboxypeptidase N subunit 1, complement component 1q subunit A, cadherin 11, paraoxonase 1, FYN binding protein 1, lysine tRNA synthetase 1, γ-tubulin complex protein 3, fibrinogen 2, GRB2-associated adaptor protein 2, Wiskott-Aldrich syndrome protein-interacting protein family member 1, ubiquitination factor homolog 6, peptidoglycan recognition protein 2, sorting connector protein 12, actin-associated protein 2 / 3 complex subunit 2, and epidermal protein 1. The blood metabolic markers include diosgenin, eicosapentaenoic acid, folinic acid, hydroxyprogesterone caproate, testosterone, perfluorooctane sulfonate, prastoratone sulfate, androstenone, pinokinone, 1-(22-carbon-4-enoyl)lysophosphatidylcholine, 16-carbon platelet-activating factor, prolylphenylalanine, testosterone sulfate, tuber acid, valine-prolyl-arginyl-aspartic acid, 5,6-dehydroarachidonic acid, 5α-androstan-3β-ol-17-ketoglucuronide, 5α-pregnane-3α,17-diol-20-one-3-sulfate, 3-carboxy-4-methyl- 5-Propyl-2-furanopropionic acid, 40:6 fatty acid esters, 12-hydroxy-9-octadecanoic acid, monoisopropyl hexadecanoic acid, branched C3 perfluoroalkyl sulfonic acid; hexafluorohexyl sulfonic acid, styromycin, (3α,5β,7α)-23-carboxy-7-hydroxy-24-choledan-3-yl-β-D-pyranoglycolic acid, 2-methyl-3-hydroxyethylidene pyran-4-one, 3-hydroxy-3',4'-dimethoxyflavone, 4-[(4-fluorophenoxy)methyl]-1-[1-[(3-methylphenyl)methyl]azacyclobutane-3-yl]triazole.

2. A detection method, characterized in that, The method is used to detect the combination of combined markers as described in claim 1, and includes the following steps: (1) Plasma sample collection and pretreatment: EDTA-anticoagulated peripheral venous blood was collected and plasma was separated by centrifugation; the plasma was then added to the complex formed by biotin-labeled polyclonal capture antibody and streptavidin magnetic beads for immunocapture, and magnetic bead complex enriched with target protein and supernatant containing metabolites were obtained. (2) Target protein processing: The magnetic bead complex obtained in step (1) is washed, eluted, denatured, reduced, alkylated and enzymatically digested to obtain protein peptide samples; (3) Treatment of target metabolites: Add pre-cooled extraction solution and internal standard mixture to the supernatant containing metabolites obtained in step (1), and then perform vortexing, static precipitation, centrifugation, concentration and reconstitution to obtain metabolite samples; (4) Simultaneous detection: Liquid chromatography-tandem mass spectrometry was used to detect the protein peptide samples in step (2) and the metabolite samples in step (3) to obtain quantitative data of each marker; (5) Risk assessment: Input the quantitative data obtained in step (4) into the machine learning model to calculate the biological age score, and determine the risk level of age-related diseases based on the biological age score.

3. The detection method according to claim 2, characterized in that, In step (1), the conditions for immune capture are 37°C and 1200 rpm shaking incubation for 60 min; The polyclonal capture antibodies include anti-APPL1, anti-ENO3, anti-P4HA1, anti-ADH1C, anti-EIF5B, anti-PMVK, anti-STARD3NL, anti-CA6, anti-CDH5, anti-IGF1, anti-CHMP4B, anti-CARHSP1, anti-COLEC11, anti-STK39, anti-MASP1, anti-CD14, anti-CETP, anti-XPOT, and anti-SSR1. Anti-MARCO, anti-IFI27, anti-SMAD1, anti-ACY1, anti-CPN1, anti-C1QA, anti-CDH11, anti-PON1, anti-FYB1, anti-KARS1, anti-TUBGCP3, anti-FCN2, anti-GRAP2, anti-WIPF1, anti-UBXN6, anti-PGLYRP2, anti-SNX12, anti-ARPC2, and anti-EPN1 were used, with each antibody administered in doses of 2 μg.

4. The detection method according to claim 2, characterized in that, In step (1), the amount of plasma used is 100 μL; the conditions for centrifuging the plasma are 4°C, 3000g centrifugation for 10 min.

5. The detection method according to claim 2, characterized in that, In step (3), the pre-cooled extract is a solution of methanol, acetonitrile and water in a volume ratio of 2:2:1; The internal standard mixture contains d4-diosgenin, d5-eicosapentaenoic acid, ¹³C5-leucovorin, d6-hydroxyprogesterone hexanoate, d3-testosterone, ¹³C8-perfluorooctane sulfonic acid, d4-praserotonin sulfate, d3-androstidyldienone, d5-pineneone, d4-1-(22C4enoyl)lysophosphatidylcholine, d3-16C hemolysophosphatidyl-activating factor, d4-prolylphenylalanine, d3-testosterone sulfate, d4-tubulic acid, ¹³C4-valine-prolyl-arginyl-aspartic acid, d5-5,6-dehydroarachidonic acid, ¹³C3-5α-androstane-3β-ol-17-ketoglucuronide, d4-5α-pregnane-3α,17-diol-20-keto-3-sulfate, and d3-3-carboxylic acid. The following internal standards were selected: d6-4-methyl-5-propyl-2-furanopropionic acid, d6-40:6 fatty acid ester, d4-12-hydroxy-9-octadecanoic acid, d5-hexadecanoic acid monoisopropyl ester, ¹³C4-branched C3 perfluoroalkyl sulfonic acid; hexafluorohexyl sulfonic acid, d4-stem-point mycotoxin, ¹³C5-(3α,5β,7α)-23-carboxy-7-hydroxy-24-cholesterol-3-yl-β-D-pyranoglycolic acid, d3-2-methyl-3-hydroxyethylidene-4-one, d4-3-hydroxy-3',4'-dimethoxyflavone, and d5-4-[(4-fluorophenoxy)methyl]-1-[1-[(3-methylphenyl)methyl]azacyclobutane-3-yl]triazole. The final concentration of each internal standard was 100 ng / mL.

6. The detection method according to claim 2, characterized in that, In step (5), the machine learning model is a random forest model or a LASSO regression model.

7. A risk prediction kit for age-related diseases, characterized in that, The detection reagent includes the combination of biomarkers according to claim 1, wherein the detection reagent comprises biotin-labeled polyclonal capture antibody, streptavidin magnetic beads, washing buffer, elution buffer, pre-cooled extraction buffer, internal standard mixture, reconstitution solution, calibrator and quality control.

8. The application of the combination of biomarkers as described in claim 1 in the preparation of an age-related disease risk prediction product.

Images