Discovering additional genetic loci associated with six psychiatric disorders/traits via FDR regression model leveraging external genetic and biological data

BackgroundCommon psychiatric disorders have substantial heritability influenced by multiple genes. While a number of susceptibility variants have been identified, many associated variants remain undiscovered. This study aimed to identify additional genetic loci associated with common psychiatric disorders/traits by leveraging correlated traits and biological annotations. MethodsWe proposed application of the false discovery rate (FDR) regression model to uncover additional genetic loci for six psychiatric disorders/traits. To enhance the likelihood of discovering additional significant genetic loci and genes, we utilized a set of 42 correlated traits and 21 biological annotations as covariates. Internal validation analysis and drug cluster enrichment analysis were conducted to validate the biological significance of the additional genetic loci/genes uncovered. We also experimentally validated two additional genes revealed for autism spectrum disorder (ASD). ResultsThe FDR regression (FDRreg) analysis strategy revealed hundreds of additional significant genes (FDR<0.01) in gene-level analyses, surpassing the number of significant genes found in the original studies. Specifically, in 11/16 trait analyses, FDRreg identified more significant genes based on gene-based analysis with MAGMA, and in 12/16 analyses, FDRreg identified more significant genes based on imputed expression in the brain. In SNP-level results, the majority of analyses (13/16) identified an equal or higher number of genomic risk loci (FDR<0.01). We found that FDRreg is able to reveal genes that are later known to be significant in subsequent larger-scale GWAS. Drug cluster enrichment analysis demonstrated a stronger enrichment in psychiatry-related drug clusters. In utero electroporation (IUE) experiments provided evidence to support two additional genes identified for ASD in critical embryonic brain development processes. ConclusionsBy integrating genetically correlated traits and biological annotations, the FDRreg strategy enables the identification of a greater number of additional significant genes and risk loci. Moreover, the new associated genes exhibited meaningful biological and clinical implications. This study presents a valuable approach for uncovering the genetic basis of psychiatric disorders and gaining insights into their underlying biology.