Association Analysis Between Genetic Variation And Multivariate Traits Based On Aggregated Dat

Genome-Wide Association Studies(GWAS)test hundreds of thousands of genetic variants in the genome to identify genetic variants that are significantly associated with specific traits or complex diseases.However,these variants explain only a small proportion of the overall heritability of most traits,and joint analysis of multiple traits can improve the power of identifying novel genetic variants,especially pleiotropic genetic variants.In addition,methods based on summary data are becoming more popular because individual-level data are difficult to obtain,whereas summary data are usually publicly available and increasingly available for various traits.In this paper,we propose the Adaptive Test based on Principal Components(ATPC),which tests the association of a single genetic variant with multiple traits based on summary data.The ATPC method first constructs a series of test statistics containing different numbers of principal components based on the trait correlation matrix,then calculates their corresponding values,and finally utilizes the Aggregated Cauchy Association Test to combine these values to obtain the ultimate test statistics.Extensive simulation studies have shown that the ATPC method can control the type I error rate,with higher power and robust performance in most scenarios compared to several multivariate trait association methods.By analyzing the summary data regarding the lipids,ATPC identified many novel SNPs that were missed by the single-trait association analysis.GO,KEGG and Reactome enrichment analysis of significantly associated SNPs detected a number of biological pathways related to lipid traits.These analyses show that the ATPC method is effective,computationally efficient,and has valuable practical applications.