Research On Identification Of Genes Related To Multiple Psychiatric Disorders Based On Multivariate Association Analysis

The revision of the classification of psychiatric illness in "The Diagnostic and Statistical Manual of Mental Disorders(DSM)" and "The International Classification of Diseases(ICD)",once again leads to the debate about the validity of the diagnostic boundary.With the the development of genome-wide association analysis study(GWAS),multivariate association analysis can provide new ways to detecte the common pathogenic genes of multiple psychiatric disorders on the molecular level.There is currently no research on gene-based public pathogenic factors for psychiatric illness.By comparing the results of multivariate association analysis with the GWAS report of single disease,we can provide important evidence for the determination of the diagnostic boundaries of psychiatric disorders.The selection of genetic correlations between diseases is a major challenge in multivariate association analysis.Based on the GWAS data among the three diseases of bipolar disorder,major depressive disorder,schizophrenia,the multivariate gene-based genome-wide association analysis(MGAS)model was used to detect the genetic correlation data provided by the mixed linear model and the LD score regression model(LD Hub).Comparing two groups of genes related to the three diseases,the results show that multivariate association analysis of genetic correlation data provided by the mixed linear model and LD Hub model have strong consistency.When it is difficult to calculate the phenotypic correlation from the original data,the genetic correlation obtained by LD Hub model can be applied to the multivariate association analysis of more types of diseases.In addition,a new multivariate association analysis data processing method is proposed in this thesis,which the number of input locus data are increased by filling in the statistical correlation test P-value of the locus.Multiple association analysis was performed on five kinds of psychiatric disorders,including Bipolar disorder,major depressive disorder,schizophrenia,attention deficit hyperactivity disorder and autism spectrum disorder.The results showed that 444 genes exceeded the genome-wide significance level,and 26 genes were in the top region of P<10-7.Comparing with the previous GWAS analysis report and the multivariate association analysis report of the five psychiatric disorders,more public pathogenic genes were detected in this work and the results were reliable.