Alzheimer's disease diagnostic marker screening method and system

Abstract

The invention relates to the field of biological medicine and data analysis, in particular to a method and a system for screening Alzheimer's disease diagnostic markers. The invention provides a method and a system for screening Alzheimer's disease diagnostic markers, which are used for searching key genes in AD pathology based on Mendel randomization (SMR) analysis of summarized data in combination with a multi-omics (risk gene methylation, expression and protein abundance information) research method, and revealing a regulation mechanism of the key genes in AD pathology. According to the method and the system provided by the invention, the diagnosis marker and the treatment target of the AD are further accurate, the understanding of a molecular mechanism in the AD disease is deepened, a new treatment target is found, an innovative strategy and a tool are provided for individualized treatment, and the life quality of a patient is finally improved.

Description

Technical Field

[0001] The present invention relates to the fields of biomedicine and data analysis, and in particular to a method and system for screening Alzheimer's disease diagnostic markers. Background Art

[0002] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention, and should not necessarily be regarded as an admission or any form of suggestion that the information constitutes the prior art already known to a person skilled in the art.

[0003] The incidence of neurodegenerative diseases in the population is increasing year by year. Neurodegenerative diseases are more common in the elderly, including Alzheimer's disease (AD), Parkinson's disease, etc. Patients will experience symptoms such as memory loss, language disorders, time and space orientation disorders, mental retardation, slow thinking, poor calculation and judgment, personality changes, and loss of initiative in life and work. In particular, AD has become the fourth disease causing high mortality in humans after cardiovascular disease, cancer, and stroke. The etiology of AD is multifactorial, including genetics, infection, traumatic brain injury, environmental and lifestyle factors, etc., but the mechanism behind each factor has not yet been elucidated, which has brought great difficulties to the discovery and treatment of the disease. Biomolecular information (such as protein, gene expression and methylation) can reflect the occurrence and development of the disease and provide information about the patient's physiological and pathological conditions. At the same time, it can also serve as an effective intervention target to provide ideas for drug development and precision treatment. Through this biomolecular information, we can have a more comprehensive understanding of the pathogenesis of AD, and at the same time, provide good conditions for the diagnosis, prediction and treatment of AD. In recent years, people have been constantly looking for more accurate biomarkers and exploring treatment options for AD. With the deepening of epigenetic research in the field of life sciences, discovering epigenetic changes in AD that are worth studying has also become a research focus.

[0004] DNA methylation is an important part of epigenetics. Currently, researchers have explored the DNA methylation process of disease genes in AD pathology, clarified the role of gene methylation modification in affecting the expression of Aβ and tau proteins, and may affect the normal expression and function of multiple genes. However, AD is a complex multifactorial disease that may involve the modification and regulation of many key genes during the onset of the disease, and the molecular regulatory network behind it remains to be elucidated. Therefore, it is crucial to further explore possible risk genes, modification processes, and their specific roles in Alzheimer's disease.

[0005] Mendelian Randomization (MR) is a genetic method that uses genetic variation as an instrumental variable to study how exposure factors affect outcome events. It can clarify the causal relationship between exposure (factor) and outcome (event). It is impractical to use randomized controlled trials to evaluate causality, and observational studies may produce biased association results due to confounding or reverse causality. MR solves these problems by using genetic variation as an instrumental variable for the exposure being studied, ensuring that the alleles of these genetic variations associated with the exposure are randomly assigned and are not affected by reverse causality.

[0006] Molecular quantitative trait loci (QTL) can be used as functional intermediates to study the underlying biological mechanisms of disease inheritance. Expression Quantitative Trait Locus (eQTL), methylation Quantitative Trait Locus (mQTL) and protein Quantitative Trait Locus (pQTL) are derived from multi-omics integrated data, revealing changes in gene expression, methylation and protein expression levels. Single Nucleotide Polymorphisms (SNPs) associated with QTLs are selected as instrumental variables (IVs) and summary-data-based Mendelian Randomization (SMR) is performed to infer the direct causal relationship between methylation, expression or protein levels of certain genes and AD. There are also studies that use pQTL to obtain markers of mental illness through two-sample Mendelian randomization methods, but it is not clear how to screen markers and therapeutic targets for Alzheimer's disease through multi-omics levels. Summary of the invention

[0007] To overcome the deficiencies of the above-mentioned prior art, the present invention provides a method and system for screening diagnostic markers for Alzheimer's disease, based on Mendelian randomization (SMR) analysis of aggregated data, combined with multi-omics (risk gene methylation, expression and protein abundance information) research methods, to find key genes in AD pathology and reveal their regulatory mechanisms in the onset of AD. The method and system provided by the present invention further refine the diagnostic markers and therapeutic targets of AD, deepen the understanding of the molecular mechanisms in AD diseases, find new therapeutic targets, provide innovative strategies and tools for individualized treatment, and ultimately improve the quality of life of patients.

[0008] In order to achieve the above technical objectives, the present invention adopts the following technical solutions:

[0009] The first aspect of the present invention provides a method for screening a diagnostic marker for Alzheimer's disease, the method comprising the following steps:

[0010] (1) Obtaining GWAS data of genetic trait variation sites (SNPs) associated with Alzheimer's disease; and simultaneously obtaining multi-omics genetic information data of a healthy population, wherein the multi-omics genetic information data includes gene methylation genetic information data, gene expression genetic information data, and protein abundance genetic information data;

[0011] (2) Analyze GWAS data and multi-omics genetic information data based on Mendelian randomization of aggregated data to obtain gene methylation markers, gene expression markers, and protein abundance markers;

[0012] (3) Performing hierarchical analysis on the omics markers obtained in step (2) to obtain diagnostic markers for Alzheimer's disease.

[0013] In one or more embodiments, the method for obtaining GWAS data of genetic trait variation sites (SNPs) associated with Alzheimer's disease includes: performing genomic variation detection, associating the genetic variation of the Alzheimer's disease population with the health data of a healthy subject population, and obtaining GWAS data of genetic trait variation sites (SNPs) associated with Alzheimer's disease.

[0014] Preferably, the genome information of the Alzheimer's disease population and the healthy subject population is obtained respectively, the reads of the sample are aligned to the reference genome, and then the variant sites are detected by SNP calling, and the genotype (genotype) of each variant site of the individual is identified by genotype calling.

[0015] In one or more embodiments, the method for obtaining gene methylome genetic information data includes:

[0016] The first batch of whole blood samples from healthy people were obtained, and the whole blood methylation was measured using the Illumina HumanMethylation450 array, and highly correlated Top Cis-mQTLs were screened.

[0017] Preferably, in the Top Cis-mQTL screening process, the screening gene interval is ±1000 kb, and the P value threshold is 5.0×10 -8 and data with a removal threshold of allele frequency difference > 0.2.

[0018] In one or more embodiments, the method for obtaining gene expression genomics genetic information data includes:

[0019] A second batch of blood and peripheral blood mononuclear cell (PBMC) samples were obtained from healthy people. The gene expression levels of the samples were analyzed by Illumina, AffyteTM U219, AffyteTM Hu-Ex v1.0 ST expression arrays and RNA-seq, and highly correlated Top Cis-eQTLs were screened.

[0020] Preferably, in the Top Cis-eQTL screening process, the screening gene interval is ±1000 kb, and the P value threshold is 5.0×10 -8 and data with a removal threshold of allele frequency difference > 0.2.

[0021] In one or more embodiments, the method for obtaining protein abundance omics genetic information data includes:

[0022] The plasma proteins of the third batch of healthy people were obtained and tested by SomaScan. GWAS was performed on the plasma levels, and highly correlated Top Cis-pQTLs were screened.

[0023] Preferably, in the Top Cis-pQTL screening process, the screening gene interval is ±1000 kb, and the P value threshold is 5.0×10 -8 and data with a removal threshold of allele frequency difference > 0.2.

[0024] In one or more embodiments, the method of analyzing GWAS data and gene methylome genetic information data based on Mendelian randomization of the aggregated data to obtain gene methylome markers includes:

[0025] Using gene methylome genetic information data as exposure factors and GWAS data as outcome factors, Mendelian randomization analysis based on summary data was performed to obtain the hypothesis test standard value P-value;

[0026] The HEIDI test was used to obtain exposure factors with HEIDI>0.01;

[0027] The Benjamini-Hochberg method was used to adjust the standard value of the hypothesis test P-value to obtain the exposure factors with the false discovery rate (FDR)-corrected P-value < 0.05;

[0028] Exposure factors with FDR-corrected P-value<0.05 and HEIDI>0.01 were screened as gene methylation biomarkers.

[0029] In one or more embodiments, the method of analyzing GWAS data and gene expression genomics genetic information data based on Mendelian randomization of the aggregated data to obtain gene expression genomics markers includes:

[0030] Using gene expression genomics genetic information data as exposure factors and GWAS data as outcome factors, Mendelian randomization analysis based on summary data was performed to obtain the hypothesis test standard value P-value;

[0031] The HEIDI test was used to obtain exposure factors with HEIDI>0.01;

[0032] The Benjamini-Hochberg method was used to adjust the standard value of the hypothesis test P-value to obtain the exposure factors with the false discovery rate (FDR)-corrected P-value < 0.05;

[0033] Exposure factors with FDR-corrected P-value<0.05 and HEIDI>0.01 were screened as gene expression markers.

[0034] In one or more embodiments, the method of analyzing GWAS data and protein abundance omics genetic information data based on Mendelian randomization of the aggregated data to obtain protein abundance omics markers includes:

[0035] Using protein abundance omics genetic information data as exposure factors and GWAS data as outcome factors, Mendelian randomization analysis based on summary data was performed to obtain the hypothesis test standard value P-value;

[0036] The HEIDI test was used to obtain exposure factors with HEIDI>0.01;

[0037] The Benjamini-Hochberg method was used to adjust the standard value of the hypothesis test P-value to obtain the exposure factors with the false discovery rate (FDR)-corrected P-value < 0.05;

[0038] Exposure factors with FDR-corrected P-value<0.05 and HEIDI>0.01 were screened as protein abundance omics markers.

[0039] In one or more embodiments, the method of performing hierarchical analysis on each omic marker obtained in step (2) to obtain a diagnostic marker for Alzheimer's disease includes:

[0040] The same markers among protein abundance omics markers, gene methylation omics markers, and gene expression omics markers are first-tier markers;

[0041] The same markers in the protein abundance omics markers and the gene methylation omics markers or the same markers in the protein abundance omics markers and the gene expression omics markers are second-tier markers.

[0042] The second aspect of the present invention provides a screening system for Alzheimer's disease diagnostic markers, comprising:

[0043] A data acquisition module is used to acquire GWAS data of genetic trait variation sites (SNPs) associated with Alzheimer's disease; and simultaneously acquire multi-omics genetic information data of healthy people, wherein the multi-omics genetic information data includes gene methylation omics genetic information data, gene expression omics genetic information data, and protein abundance omics genetic information data;

[0044] Data analysis module, used to analyze GWAS data and multi-omics genetic information data based on Mendelian randomization of aggregated data, and obtain gene methylation omics markers, gene expression omics markers, and protein abundance omics markers;

[0045] The marker acquisition module is used to perform hierarchical analysis on the various omics markers obtained in the data analysis module to obtain diagnostic markers for Alzheimer's disease.

[0046] The third aspect of the present invention provides an Alzheimer's disease diagnostic marker obtained by screening by the screening method of the first aspect and / or the screening system of the second aspect;

[0047] The Alzheimer's disease diagnostic markers include: TMEM106B, IBSP, FCRLB, SIGLEC9, CD33, LILRB1, PLIRA, ACE, GRN, NSF, SIRPA, LILRA4, UBASH3B or LY6G6D.

[0048] The beneficial effects of the present invention are:

[0049] The present invention uses genetic information data of three omics, namely gene methylation, gene expression and protein abundance level of healthy people and genetic information data of Alzheimer's disease to conduct Mendelian randomization analysis based on aggregated data, and finds key genes and corresponding single nucleotide polymorphisms in Alzheimer's disease, providing important reference for the prediction, diagnosis and intervention treatment of the course of Alzheimer's disease. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0051] Figure 1 This is a forest plot of the SMR results of gene methylation and Alzheimer's disease in Example 1 of the present invention;

[0052] Figure 2 A to F in the figure are forest plots of the SMR results of genes and Alzheimer's disease in Example 1 of the present invention;

[0053] Figure 3 This is a forest plot of the SMR results of protein and Alzheimer's disease in Example 1 of the present invention;

[0054] Figure 4 It is the MR forest map of the corresponding methylation sites and ACE gene and CD33 gene in the first-level evidence in Example 1 of the present invention;

[0055] Figure 5 It is the result of integrating multi-omics evidence in Example 1 of the present invention;

[0056] Figure 6 The results of the database verification in Example 2 of the present invention; wherein A is the verification of the gene methylation level, B and C are the verification of the gene expression level, and D is the verification of the plasma protein level;

[0057] Figure 7 A and B are the differential expressions of important risk genes in AD and healthy people as described in Example 3 of the present invention;

[0058] Figure 8 A and B are the sensitivity and specificity analysis results of important risk genes described in Example 3 of the present invention. DETAILED DESCRIPTION

[0059] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0060] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0061] In order to enable those skilled in the art to more clearly understand the technical solution of the present invention, the technical solution of the present invention will be described in detail below in conjunction with specific embodiments.

[0062] Example 1

[0063] This embodiment provides a method for screening Alzheimer's disease diagnostic markers, the method comprising the following steps:

[0064] (1) Obtaining GWAS data of genetic trait variation sites (SNPs) associated with Alzheimer's disease; and simultaneously obtaining multi-omics genetic information data of a healthy population, wherein the multi-omics genetic information data includes gene methylation genetic information data, gene expression genetic information data, and protein abundance genetic information data;

[0065] (2) Analyze GWAS data and multi-omics genetic information data based on Mendelian randomization of aggregated data to obtain gene methylation markers, gene expression markers, and protein abundance markers;

[0066] (3) Performing hierarchical analysis on the omics markers obtained in step (2) to obtain diagnostic markers for Alzheimer's disease.

[0067] In step (1), the method for obtaining GWAS data of genetic trait variation sites (SNPs) associated with Alzheimer's disease is:

[0068] By performing genomic variation detection, the genetic variations of the Alzheimer's disease population are associated with the health data of the healthy subject population, and GWAS data of genetic trait variation sites (SNPs) related to Alzheimer's disease are obtained.

[0069] Specifically, the genetic variation of the Alzheimer's disease population and the healthy subjects came from the European Alzheimer & Dementia Biobank (EADB) Consortium, with a total of 85,934 AD patients and 401,577 healthy subjects. The genome information of these two populations was obtained, the reads of the samples were aligned to the reference genome, and then the variant sites were detected by SNP calling, and the genotype of each variant site of the individual was identified by genotype calling.

[0070] In step (1), the method for obtaining gene methylome genetic information data includes:

[0071] The first batch of whole blood samples from healthy people were obtained, and the whole blood methylation was measured using the Illumina HumanMethylation450 array, and highly correlated Top Cis-mQTLs were screened.

[0072] Specifically:

[0073] The first batch of healthy population samples came from the following databases (Brisbane Systems Genetics Study (BSGS) and Lothian Birth Cohorts (LBC)), with a total of 1,980 subjects. Whole blood samples were obtained from the subjects, and whole blood methylation was measured using the Illumina HumanMethylation450 array, obtaining 52,916 cis-DNA methylation quantitative trait loci (mQTLs).

[0074] Based on the collected mQTL data, the screening gene interval was ±1000 kb, and the P value threshold was 5.0×10 -8 And remove the data with a threshold of allele frequency difference > 0.2 as highly associated Top Cis-mQTL. In this process, SNPs with an allele frequency difference greater than the specified threshold (the specified threshold is 0.2) between any paired datasets were excluded.

[0075] In step (1), the method for obtaining the genetic information data of gene expression genomics is:

[0076] A second batch of blood and peripheral blood mononuclear cell (PBMC) samples were obtained from healthy people and preprocessed in a standardized manner. The gene expression levels of the samples were analyzed by Illumina, AffyteTM U219, AffyteTM Hu-Ex v1.0 ST expression arrays and RNA-seq, and highly correlated Top Cis-eQTLs were screened.

[0077] Specifically:

[0078] The first batch of healthy population samples came from the eQTLGen Consortium database, with a total of 31,684 subjects. Blood and peripheral blood mononuclear cell (PBMC) samples were obtained and preprocessed in a standardized manner. The gene expression levels of the samples were analyzed by Illumina, AffyteTM U219, AffyteTM Hu-Ex v1.0 ST expression arrays and RNA-seq. Genotype and expression data preprocessing, PGS calculation, and cis-eQTL, trans-eQTL and eQTL mapping were performed. The standardized preprocessing included genotype inference, data coordination, SNP screening, and identification using a specific database version (using a unified genome version, if the data is GRCh37 / hg19, use the GRCh37 genome version for identification. If it is GRCh38 / hg38, use the GRCh38 genome version for identification), to ensure the accuracy and reliability of the study.

[0079] The screening gene interval was ±1000 kb, and the P value threshold was 5.0×10 -8 And the data with a removal threshold of allele frequency difference > 0.2 were highly correlated Top Cis-eQTL.

[0080] In step (1), the method for obtaining protein abundance omics genetic information data includes:

[0081] The plasma proteins of the third batch of healthy people were obtained and tested by SomaScan. GWAS was performed on the plasma levels, and highly correlated Top Cis-pQTLs were screened.

[0082] Specifically:

[0083] The third batch of healthy people from the Icelandic Cancer Project and deCODE genetics, Reykjavík, Iceland project databases were obtained, with a total of 35,559 subjects. Plasma proteins were extracted and separated. SomaScan was used to detect 27 million variants and plasma levels of 4,907 adapters. GWAS was performed, and 28,191 pQTL associations were identified, of which 27% were cis, 73% were trans, and 93% were unreported. pQTLs were associated with the levels of one or more proteins.

[0084] The data with a screening gene interval of ±1000 kb, a P value threshold of 5.0×10-8, and a removal threshold of allele frequency difference >0.2 were highly correlated Top Cis-pQTL.

[0085] In step (2), the method of analyzing the GWAS data and the gene methylome genetic information data based on the Mendelian randomization of the aggregated data to obtain the gene methylome markers includes:

[0086] Using gene methylome genetic information data as exposure factors and GWAS data as outcome factors, Mendelian randomization analysis based on summary data was performed to obtain the hypothesis test standard value P-value;

[0087] The HEIDI test was used to obtain exposure factors with HEIDI>0.01;

[0088] The Benjamini-Hochberg method was used to adjust the standard value of the hypothesis test P-value to obtain the exposure factors with the false discovery rate (FDR)-corrected P-value < 0.05;

[0089] Exposure factors with FDR-corrected P-value<0.05 and HEIDI>0.01 were screened as gene methylation biomarkers.

[0090] In step (2), the method of analyzing the GWAS data and the gene expression genomics genetic information data based on the Mendelian randomization of the aggregated data to obtain the gene expression genomics markers includes:

[0091] Using gene expression genomics genetic information data as exposure factors and GWAS data as outcome factors, Mendelian randomization analysis based on summary data was performed to obtain the hypothesis test standard value P-value;

[0092] The HEIDI test was used to obtain exposure factors with HEIDI>0.01;

[0093] The Benjamini-Hochberg method was used to adjust the standard value of the hypothesis test P-value to obtain the exposure factors with the false discovery rate (FDR)-corrected P-value < 0.05;

[0094] Exposure factors with FDR-corrected P-value<0.05 and HEIDI>0.01 were screened as gene expression markers.

[0095] In step (2), the method of analyzing the GWAS data and the protein abundance omics genetic information data based on the Mendelian randomization of the aggregated data to obtain the protein abundance omics markers includes:

[0096] Using protein abundance omics genetic information data as exposure factors and GWAS data as outcome factors, Mendelian randomization analysis based on summary data was performed to obtain the hypothesis test standard value P-value;

[0097] The HEIDI test was used to obtain exposure factors with HEIDI>0.01;

[0098] The Benjamini-Hochberg method was used to adjust the standard value of the hypothesis test P-value to obtain the exposure factors with the false discovery rate (FDR)-corrected P-value < 0.05;

[0099] Exposure factors with FDR-corrected P-value<0.05 and HEIDI>0.01 were screened as protein abundance omics markers.

[0100] In the above step (2), the formula for obtaining the exposure factor with HEIDI>0.01 using the HEIDI test is:

[0101]

[0102] The subscript “0” indicates the causal variant, the subscript “i” indicates SNP i, and the 0i is the linkage disequilibrium association between the causal variant and SNP i, and h = 2p(1-p), where p is the allele frequency. Therefore, a linkage test for pleiotropy (two different causal variants, one affecting gene expression and one affecting the trait) is equivalent to testing b xx(top) (b estimated using top associated cis-eQTL xy ) and b xy(i) (b estimated using any other significant SNPs in the cis-eQTL region xy ) is there a difference between them? If we define d i =b xy(i) —b xy(top) , is further equivalent to testing all significant cis-eQTLs (excluding top cis-eQTLs) i Is it equal to 0. If we define where m is the number of significant eQTLs (excluding top cis-eQTLs), Here V is the (co)variance matrix, where the ijth element is

[0103]

[0104] and The covariance between is calculated as:

[0105]

[0106] where r ij is the linkage disequilibrium correlation between SNPs i and j. Under the null hypothesis that there is no heterogeneity, i.e., d = 0, the standard normal vector is represented by z d ={z d(1) ,z d(2) ,…,z d(m)}, z d ~MVN(0,R), where R is the correlation matrix, can be expressed as To test for d=0, construct the HEIDI test Right now where I is the identity matrix.

[0107] HEIDI test was used to distinguish true conjoined effects from out-of-conjoined effects. For cases with P-HEIDI < 0.01, they were considered out-of-conjoined effects and therefore excluded from the analysis. This process ensured the reliability and accuracy of our analysis results.

[0108] In the above step (2), the method of adjusting the hypothesis test standard value P-value using the Benjamini-Hochberg method is:

[0109] ① Sort all P-values ​​from small to large;

[0110] ② Assign a ranking to the P-value: the smallest P-value is ranked 1, the second smallest is ranked 2, and so on;

[0111] ③ Calculate the BH critical value corresponding to each P-value using the formula: (i / m)Q; where i is the rank of the p-value, m is the total number of tests, and Q is the false positive rate (FDR, a percentage chosen by the researcher)

[0112] ④ Compare the original P-value and the BH critical value calculated in step 3: find the largest P-value that is "smaller than its corresponding BH critical value".

[0113] The method for correcting the false discovery rate (FDR) is:

[0114] Sort the results of a total of m tests from small to large, and k is the ranking corresponding to the P-value of one of the test results.

[0115] Find the largest k value that meets the original threshold α, satisfying P(k)<=α*k / m, and consider that all tests ranked from 1 to k have significant differences, and calculate the corresponding q value formula as q=p*(m / k).

[0116] In the above step (2), the GWAS data and multi-omics genetic information data were analyzed based on Mendelian randomization of the aggregated data to obtain gene methylation omics markers, gene expression omics markers, and protein abundance omics markers; colocalization analysis was performed using the "coloc" R package to determine whether two or more phenotypes (such as diseases or molecular phenotypes) were affected by the same causal variant site in a specific region. Colocalization analysis produces four situations: (1) The two phenotypes are not significantly associated with all SNP sites in the genomic region (H0); (2) One of the phenotypes is significantly associated with a SNP site in a certain genomic region (H1); (3) Both phenotypes are significantly associated with a SNP site in a certain genomic region, but are driven by different causal variant sites (H3); (4) Both phenotypes are significantly associated with a SNP site in a certain genomic region and are driven by the same causal variant site (H4). Among them, for the colocalization of mQTL-GWAS, the colocalization region window is ±500kb. The prior probabilities that a SNP is associated with trait 1 (mQTL), trait 2 (AD), and both traits are associated are set to 10 -4 , 10 -4 and 10 -5 In our study, a posterior probability H4 (PP.H4) greater than 0.5 indicated successful colocalization, suggesting that the two phenotypes were driven by the same causal variant. For the colocalization of eQTL-GWAS, the colocalization region window was ±1000 kb. The prior probabilities that a SNP was associated with trait 1 (eQTL), trait 2 (AD), and both traits were associated were set to 10, respectively. -4 , 10 -4 and 10 -5 In our study, a posterior probability H4 (PP.H4) greater than 0.5 indicated successful colocalization, suggesting that the two phenotypes were driven by the same causal variant. For colocalization in pQTL-GWAS, the colocalization region window was ±1000 kb. The prior probabilities that a SNP was associated with trait 1 (pQTL), trait 2 (AD), or both traits were set to 10, respectively. -4 , 10 -4 and 10 -5 In our study, a posterior probability H4 (PP.H4) greater than 0.5 indicated successful colocalization, suggesting that the two phenotypes were driven by the same causal variant.

[0117] result:

[0118] Visualize causal gene methylation information through forest plot: Use the "forestploter" package in the R program to visualize gene methylation and related information that have a strong causal relationship with AD (P<0.05). Specific information includes methylation sites, genes, SNP sites, forest plots, odds ratios (OR), 95% confidence intervals, hypothesis test standard values ​​P-value, and co-localization results PP.H4. Finally, 39 CpG sites (25 genes) were found to be related to the development of AD (Table 1). Among them, multiple CpG sites in the same gene can be used to predict that the gene is related to the risk of AD, for example, methylation of 24 CpG sites on gene TMEM184A (cg00411097, cg01726448, cg02879083, cg03564557, cg03664589, cg03723321, cg03958363, cg04501217, cg04633141, cg05868365 ,cg06194808,cg07071809,cg08729755,cg09329079,cg09495282,cg09990600,cg10633906,cg12503394,cg14480560,cg18294109,cg18648645,cg18705808,cg21368161,cg23404435) were found to increase or decrease the risk of AD. According to the results of the OR value, different gene methylation may increase or decrease the risk of AD. For some genes, the methylation predicted at different CpG sites may play opposite roles in the development of the disease. For example, for the RSPO4 gene, methylation at cg00862894 is predicted to increase the risk of AD, while tests at cg03569404, cg04199533, and cg26278454 found that methylation of RSPO4 would reduce the risk of disease ( Figure 1 ).

[0119] Table 1: AD markers based on gene methylome

[0120]

[0121]

[0122] Visualize causal gene expression information through forest plot: Use the "forestploter" package in the R program to visualize genes with strong causal relationship (P<0.05) with AD and related information. Specific information includes related genes, SNP sites, forest plots, odds ratios (OR), 95% confidence intervals, and hypothesis test standard values ​​P-values.

[0123] A total of 125 genes were found to have a causal relationship between the expression level of the gene and the onset of AD. It was found that the increased expression level of 61 genes would increase the risk of AD (OR>1), and the high expression level of the remaining 64 genes might reduce the risk of AD ( Figure 2 ), among which 125 genes and loci are shown in Table 2:

[0124] Table 2: AD markers at genomic expression levels

[0125]

[0126]

[0127]

[0128] Visualize the causal protein expression information through forest plot: Use the "forestploter" package in the R program to visualize the abundance of proteins with a strong causal relationship (P<0.05) with AD and related information. Specific information includes related proteins, SNP sites, forest plots, odds ratios (OR), 95% confidence intervals, and hypothesis test standard values ​​P-value.

[0129] Finally, 18 AD risk-related proteins were obtained (Table 3). According to genetic prediction, high expression levels of 8 proteins will increase the risk of AD, including TMEM106B (OR 1.44CI 1.24, 1.68), IBSP (OR 1.35CI 1.13, 1.61), FCRLB (OR 1.11CI 1.04, 1.17), BPNT2 (OR 1.11CI 1.05, 1.17), CTSH (OR 1.04CI1.03, 1.06), SIGLEC9 (OR 1.03CI 1.02, 1.04), CD33 (OR 1.03CI 1.01, 1.05), SIRPA (OR1.03CI 1.02, 1.04). Increased expression levels of 10 proteins are associated with a reduced prevalence of AD, including LILRB1 (OR 0.95CI 0.93, 0.97), PILRA (OR 0.94CI 0.93,0.96),ACE(OR 0.92CI 0.89,0.94),LGALS3(OR 0.91CI 0.87,0.96),LILRA4(OR 0.9CI 0.85,0.95),UBASH3B(OR 0.77CI 0.66,0.88),GRN(OR 0.74CI 0.68,0.81),CLN5(OR 0.69CI 0.58,0.83),NSF(OR 0.6CI 0.5,0.72),LY6G6D(OR 0.27CI 0.15,0.48)( Figure 3 ).

[0130] Table 3: 18 AD risk-associated proteins

[0131]

[0132] In step (3), in order to further improve the accuracy, the SMR data of different omics and AD are collated to form different levels of evidence. Through the previous SMR study, the correlation between gene methylation, gene expression, and the abundance level of proteins obtained after gene translation and the onset of AD has been obtained. That is, from the perspective of the three omics, it is explained that there is a causal relationship between the expression changes of certain genes and the pathological progression of AD. The regulation and expression of genes is a complex and continuous process. If you want to prove that a certain gene itself does play a key role in the pathological progression of the disease, you must form a complete chain of evidence. Each piece of evidence is obtained by merging data results with statistical differences (SMR1: gene methylation and AD; SMR2: gene expression level and AD; SMR3: protein abundance level and AD). The more rigorous the data acquisition process to support the evidence, the more convincing the evidence is.

[0133] Therefore, the causal relationships obtained were sorted out and summarized into two different echelons of evidence. Protein is the basic functional unit and has the most direct relationship with disease, so the causal relationship between protein abundance level and disease is regarded as the most basic common condition.

[0134] SMR is used to determine the causal relationship between gene methylation and gene expression, and between gene expression and protein abundance, to infer how gene methylation affects gene expression, and to supplement the evidence obtained.

[0135] Results: The same markers in protein abundance omics markers, gene methylation omics markers and gene expression omics markers were the first-tier markers;

[0136] The same markers in the protein abundance omics markers and the gene methylation omics markers or the same markers in the protein abundance omics markers and the gene expression omics markers are second-tier markers.

[0137] First-tier evidence: There is a high degree of causal relationship between protein abundance level and AD (FDR-corrected P-value<0.05and P-HEIDI>0.01) and there is a high degree of causal relationship between gene methylation and gene expression level and AD (FDR-corrected P-value<0.05and P-HEIDI>0.01);

[0138] The second tier of evidence: There is a high degree of causal relationship between protein abundance level and AD (FDR-corrected P-value<0.05and P-HEIDI>0.01) and there is a high degree of causal relationship between gene methylation or gene expression level and AD (FDR-corrected P-value<0.05and P-HEIDI>0.01).

[0139] By integrating and grading multi-omics evidence, the persuasiveness and accuracy of some evidence can be strengthened, and its potential molecular mechanism can be preliminarily speculated ( Figure 5 ). The causal relationship between the two key genes ACE and CD33 and AD is classified as first-level evidence, because these two genes are causally related to the progression of AD in terms of methylation, expression level, and related protein abundance level. This causal relationship meets the multiple correction criteria and excludes the in vitro effect in the HEIDI verification. At the same time, the co-localization effect with AD at the three omics levels is good, which is the most convincing and can be used as a key research target ( Figure 4 ).

[0140] The causal relationship between the levels of proteins related to the five genes CTSH, SIRPA, TMEM106B, FCRLB, and UBASH3B and the progression of AD can be proved by SMR (FDR-corrected P-value < 0.05 and P-HEIDI > 0.01), which meets the most basic premise of protein-function influence. However, due to the lack of evidence of the corresponding methylation or gene expression level, it is classified as Level 2 evidence ( Figure 6 ).

[0141] Increased CLN5 protein levels may be associated with a reduced risk of AD (OR 0.96CI 0.58, 0.83). Due to the lack of evidence of gene expression or gene methylation levels, it cannot be classified as level 1 or 2 evidence, but the co-localization results between its protein level and AD indicate that the causal relationship between the two is driven by the same variant site, strengthening the causal relationship between CLN5 protein level and AD. Therefore, the key role of CLN5 in AD should not be ignored and has certain research value.

[0142] The discovery of the above-mentioned key genes, methylation sites and corresponding SNPs provides ideas for clarifying the molecular mechanism of AD disease and has great research significance.

[0143] Example 2

[0144] In order to verify the feasibility of the markers obtained above, this embodiment provides a method for verifying the markers obtained in three aspects using a database.

[0145] Exposure data: mQTL (methylation sequences obtained by McRae et al.), eQTL (obtained from the eQTLGen consortium), pQTL (protein quantitative trait locus study by Ferkingstad et al.); outcome data: The FinnGenStudy G6_AD_WIDE. There was no overlap in samples between the exposure population and the outcome population.

[0146] Specifically,

[0147] (1) Obtain the verification data required for this study:

[0148] AD data: The validation data for AD comes from the Finnish database (Kurki, MI, Karjalainen, J., Palta, P. et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature 613, 508–518 (2023). https: / / doi.org / 10.1038 / s41586-022-05473-8). The FinnGen study is a large-scale genomics project that has analyzed more than 500,000 Finnish biobank samples and associated genetic variations with health data to understand disease mechanisms and susceptibility. The dataset used in this study is "G6_AD_WIDE", and the genetic variation information in this dataset comes from 15,617 AD patients and 396,564 healthy subjects.

[0149] The mQTL data were derived from the methylation sequences that had been sequenced by McRae et al.

[0150] The eQTL data were obtained from the eQTLGen consortium.

[0151] The pQTL data come from a study of protein quantitative trait loci (pQTL) on individuals from Iceland by Ferkingstad et al.

[0152] (2) The same processing method as in Example 1 was used to obtain gene methylation omics markers, gene expression omics markers and protein abundance omics markers.

[0153] result:

[0154] Visualize causal molecular information through forest plots: Use the "forestploter" package in the R program to visualize molecular information that has a strong causal relationship with AD (P<0.05). Specific information includes methylation sites, gene expression, related proteins, SNP sites, forest plots, odds ratios (OR), 95% confidence intervals, and hypothesis test standard values ​​P-values.

[0155] At the methylation level, the causal relationship between 4 CpG sites (3 genes) and AD was successfully verified; at the gene expression level, the causal relationship between 5 genes and AD was successfully verified; at the protein abundance level, the causal relationship between 5 plasma protein levels and AD was successfully verified.

[0156] At the methylation level, 12 CpG sites (11 genes) were found to be associated with the development of AD. At the gene level, a total of 52 genes were found to have a causal relationship between the expression level of the gene and the onset of AD. We found that increased expression levels of 19 genes would increase the risk of AD (OR>1), and high levels of expression of the remaining 33 genes may reduce the risk of AD. The causal relationship between 4 CpG sites (3 genes) and AD was successfully verified, namely cg26167930 (DAPK2), cg00443543 (SERPINF2), cg18278729 (SERPINF2), cg23422036 (TMEM106B) ( Figure 6 A). The causal relationship between five genes and AD has been successfully verified, namely GRINA, RITA1, APH1B, ACE, and SIGLEC18P ( Figure 6 B). At the protein abundance level, 14 AD risk-related proteins were finally obtained, of which 5 proteins were verified to have a causal relationship with AD, namely PILRA, GRN, ACE, SIGLEC9, and CD33 ( Figure 6 C).

[0157] Example 3

[0158] In order to further determine the diagnostic efficacy of the above markers, ROC analysis was performed.

[0159] Data from postmortem temporal cortical brain samples from 71 individuals investigated in a research project at the Neuro Center, Kuopio University Hospital were collected, including information such as expression matrices and clinical samples. The data included expression information of transcripts, proteins, and phosphopeptides from temporal lobe tissues of 54 AD patients and 6 normal subjects. After grouping the data and annotating the probes, the expression of 17 genes in 60 samples, including TMEM106B, IBSP, FCRLB, CTSH, SIGLEC9, CD33, LILRB1, PILRA, ACE, LGALS3, GRN, CLN5, NSF, SIRPA, LILRA4, UBASH3B, and LY6G6D, was selected for subsequent analysis (BPNT2 expression was not present in this dataset, so it was ignored). We found that the expression of some genes in the two different populations was different through the Wilcox test, and these differences were visualized through the R package "ggplot", and the R package "pROC" was used to draw the ROC curve and calculate the AUC.

[0160] The expressions of six proteins, CD33, LILRA4, NSF, PILRA, SIRPA, and TMEM106B, were significantly different between AD and healthy subjects ( Figure 7 ), and it was found that the AUC of 17 genes were all above 0.5, indicating that these genes have certain diagnostic performance. Among them, TMEM106B, IBSP, FCRLB, SIGLEC9, CD33, LILRB1, PLIRA, ACE, GRN, NSF, SIRPA, LILRA4, UBASH3B, LY6G6D, a total of 14 genes, had an AUC of more than 0.6, with high accuracy ( Figure 8 ).

[0161] Example 4

[0162] A screening system for Alzheimer's disease diagnostic markers, comprising:

[0163] A data acquisition module is used to acquire GWAS data of genetic trait variation sites (SNPs) associated with Alzheimer's disease; and simultaneously acquire multi-omics genetic information data of healthy people, wherein the multi-omics genetic information data includes gene methylation omics genetic information data, gene expression omics genetic information data, and protein abundance omics genetic information data;

[0164] Data analysis module, used to analyze GWAS data and multi-omics genetic information data based on Mendelian randomization of aggregated data, and obtain gene methylation omics markers, gene expression omics markers, and protein abundance omics markers;

[0165] The marker acquisition module is used to perform hierarchical analysis on the various omics markers obtained in the data analysis module to obtain diagnostic markers for Alzheimer's disease.

[0166] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for screening Alzheimer's disease diagnostic markers, characterized in that: The method comprises the following steps: (1) Obtaining GWAS data of genetic trait variation sites associated with Alzheimer's disease; and simultaneously obtaining multi-omics genetic information data of healthy people, wherein the multi-omics genetic information data includes gene methylation genetic information data, gene expression genetic information data, and protein abundance genetic information data; (2) Analyze GWAS data and multi-omics genetic information data based on Mendelian randomization of aggregated data to obtain gene methylation markers, gene expression markers, and protein abundance markers; (3) Performing hierarchical analysis on the omics markers obtained in step (2) to obtain diagnostic markers for Alzheimer's disease.

2. The screening method according to claim 1, characterized in that The method for obtaining GWAS data of genetic trait variation sites associated with Alzheimer's disease includes: performing genome variation detection to associate genetic variations of a population of Alzheimer's disease patients with health data of a population of healthy subjects, thereby obtaining GWAS data of genetic trait variation sites associated with Alzheimer's disease; Preferably, the genome information of the Alzheimer's disease population and the healthy subject population is obtained respectively, the reads of the sample are aligned to the reference genome, and then the variant sites are detected by SNP calling, and the genotype of each variant site of the individual is identified by genotype calling.

3. The screening method according to claim 1, characterized in that Methods for obtaining genetic information data of gene methylome include: The first batch of healthy people's whole blood samples were obtained, and the whole blood methylation was measured using the Illumina HumanMethylation450 array, and highly correlated Top Cis-mQTLs were screened; Preferably, in the Top Cis-mQTL screening process, the screening gene interval is ±1000 kb, and the P value threshold is 5.0×10 -8 and data with a removal threshold of allele frequency difference > 0.

2.

4. The screening method according to claim 1, wherein Methods for obtaining genetic information data of gene expression genomics include: A second batch of blood and peripheral blood mononuclear cell samples were obtained from healthy people. The gene expression levels of the samples were analyzed by Illumina, AffyteTM U219, AffyteTM Hu-Ex v1.0 ST expression arrays and RNA-seq, and highly correlated Top Cis-eQTLs were screened. Preferably, in the Top Cis-eQTL screening process, the screening gene interval is ±1000 kb, and the P value threshold is 5.0×10 -8 and data with a removal threshold of allele frequency difference > 0.

2.

5. The screening method according to claim 1, characterized in that Methods for obtaining proteomic genetic information data include: The plasma proteins of the third batch of healthy people were obtained, tested by SomaScan, and GWAS was performed on the plasma levels to screen out the highly correlated Top Cis-pQTLs; Preferably, in the Top Cis-pQTL screening process, the screening gene interval is ±1000 kb, and the P value threshold is 5.0×10 -8 and data with a threshold of allele frequency difference > 0.2 were removed.

6. The screening method according to claim 1, characterized in that The method of analyzing GWAS data and gene methylome genetic information data based on Mendelian randomization of aggregated data to obtain gene methylome markers includes: Using gene methylome genetic information data as exposure factors and GWAS data as outcome factors, a Mendelian randomization analysis based on the summary data was performed to obtain the P-value. The HEIDI test was used to obtain exposure factors with HEIDI>0.01; The Benjamini-Hochberg method was used to adjust the standard value P-value of the hypothesis test to obtain the exposure factors with the false discovery rate-corrected FDR-corrected P-value < 0.05; Exposure factors with FDR-corrected P-value<0.05 and HEIDI>0.01 were screened as gene methylation biomarkers.

7. The screening method according to claim 1, wherein The method of analyzing GWAS data and gene expression genomics genetic information data based on Mendelian randomization of aggregated data to obtain gene expression genomics markers includes: Using gene expression genomics genetic information data as exposure factors and GWAS data as outcome factors, Mendelian randomization analysis based on summary data was performed to obtain the hypothesis test standard value P-value; The HEIDI test was used to obtain exposure factors with HEIDI>0.01; The Benjamini-Hochberg method was used to adjust the standard value P-value of the hypothesis test to obtain the exposure factors with the false discovery rate-corrected FDR-corrected P-value < 0.05; Exposure factors with FDR-corrected P-value < 0.05 and HEIDI > 0.01 were screened as gene expression omics markers; Alternatively, the method of analyzing GWAS data and protein abundance omics genetic information data based on Mendelian randomization of the aggregated data to obtain protein abundance omics markers includes: Using protein abundance omics genetic information data as exposure factors and GWAS data as outcome factors, Mendelian randomization analysis based on summary data was performed to obtain the hypothesis test standard value P-value; The HEIDI test was used to obtain exposure factors with HEIDI>0.01; The Benjamini-Hochberg method was used to adjust the standard value P-value of the hypothesis test to obtain the exposure factors with the false discovery rate-corrected FDR-corrected P-value < 0.05; Exposure factors with FDR-corrected P-value<0.05 and HEIDI>0.01 were screened as protein abundance omics markers.

8. The screening method according to claim 1, characterized in that The method of performing hierarchical analysis on each omic marker obtained in step (2) to obtain a diagnostic marker for Alzheimer's disease includes: The same markers among protein abundance omics markers, gene methylation omics markers, and gene expression omics markers are first-tier markers; The same markers in the protein abundance omics markers and the gene methylation omics markers or the same markers in the protein abundance omics markers and the gene expression omics markers are second-tier markers.

9. A screening system for Alzheimer's disease diagnostic markers, characterized in that: include: A data acquisition module is used to acquire GWAS data of genetic trait variation sites related to Alzheimer's disease; and simultaneously acquire multi-omics genetic information data of healthy people, wherein the multi-omics genetic information data includes gene methylation omics genetic information data, gene expression omics genetic information data, and protein abundance omics genetic information data; Data analysis module, used to analyze GWAS data and multi-omics genetic information data based on Mendelian randomization of aggregated data, and obtain gene methylation omics markers, gene expression omics markers, and protein abundance omics markers; The marker acquisition module is used to perform hierarchical analysis on the various omics markers obtained in the data analysis module to obtain Alzheimer's disease diagnostic markers.

10. A diagnostic marker for Alzheimer's disease, characterized in that: Obtained by screening according to any one of the screening methods of claims 1 to 8 and / or the screening system of claim 9; The Alzheimer's disease diagnostic markers include: TMEM106B, IBSP, FCRLB, SIGLEC9, CD33, LILRB1, PLIRA, ACE, GRN, NSF, SIRPA, LILRA4, UBASH3B or LY6G6D.

Images