Abstract:Genome-wide association studies (GWAS) of psychiatric disorders (PD) yield numerous loci with significant signals, but often do not implicate specific genes. Because GWAS risk loci are enriched in expression/protein/methylation quantitative loci (e/p/mQTL, hereafter xQTL), transcriptome/proteome/methylome-wide association studies (T/P/MWAS, hereafter XWAS) that integrate xQTL and GWAS information, can link GWAS signals to effects on specific genes. To further increase detection power, gene signals are aggregated within relevant gene sets (GS) by performing gene set enrichment (GSE) analyses. Often GSE methods test for enrichment of “signal” genes in curated GS while overlooking their linkage disequilibrium (LD) structure, allowing for the possibility of increased false positive rates. Moreover, no GSE tool uses xQTL information to perform mendelian randomization (MR) analysis. To make causal inference on association between PD and GS, we develop a novel MR GSE (MR-GSE) procedure. First, we generate a “synthetic” GWAS for each MSigDB GS by aggregating summary statistics for x-level (mRNA, protein or DNA methylation (DNAm) levels) from the largest xQTL studies available) of genes in a GS. Second, we use synthetic GS GWAS as exposure in a generalized summary-data-based-MR analysis of complex trait outcomes. We applied MR-GSE to GWAS of nine important PD. When applied to the underpowered opioid use disorder GWAS, none of the four analyses yielded any signals, which suggests a good control of false positive rates. For other PD, MR-GSE greatly increased the detection of GO terms signals (2,594) when compared to the commonly used (non-MR) GSE method (286). Some of the findings might be easier to adapt for treatment, e.g., our analyses suggest modest positive effects for supplementation with certain vitamins and/or omega-3 for schizophrenia, bipolar and major depression disorder patients. Similar to other MR methods, when applying MR-GSE researchers should be mindful of the confounding effects of horizontal pleiotropy on statistical inference.

Systematic comparative analysis of Mendelian randomization methods for inferring causal genes of complex phenotypes and the application to psychiatric diseases

Mendelian randomization: causal inference leveraging genetic data

Powerful and robust inference of complex phenotypes' causal genes with dependent expression quantitative loci by a median-based Mendelian randomization.

A robust two-sample Mendelian Randomization method integrating GWAS with multi-tissue eQTL summary statistics

A novel multi-omics mendelian randomization method for gene set enrichment and its application to psychiatric disorders

Mendelian randomization accounting for correlated and uncorrelated pleiotropic effects using genome-wide summary statistics

SMR-Portal: an online platform for integrative analysis of GWAS and xQTL data to identify complex trait genes

A robust two‐sample transcriptome‐wide Mendelian randomization method integrating GWAS with multi‐tissue eQTL summary statistics

Mendelian randomization for causal inference accounting for pleiotropy and sample structure using genome-wide summary statistics

PsyRiskMR: a Comprehensive Resource for Identifying Psychiatric Disorders Risk Factors Through Mendelian Randomization

Genome-wide Mendelian randomization identifies actionable novel drug targets for psychiatric disorders

Mendelian randomization accounting for complex correlated horizontal pleiotropy while elucidating shared genetic etiology

Bayesian Weighted Mendelian Randomization for Causal Inference based on Summary Statistics

Benchmarking Mendelian Randomization methods for causal inference using genome-wide association study summary statistics

Likelihood-based Mendelian randomization analysis with automated instrument selection and horizontal pleiotropic modeling

Reciprocal causation mixture model for robust Mendelian randomization analysis using genome-scale summary data

MR-link-2: pleiotropy robust cis Mendelian randomization validated in four independent gold-standard datasets of causality

An integrative multi-context Mendelian randomization method for identifying risk genes across human tissues

Mendelian randomization under the omnigenic architecture

Mendelian Randomization With Refined Instrumental Variables From Genetic Score Improves Accuracy and Reduces Bias

MR-LDP: a Two-Sample Mendelian Randomization for GWAS Summary Statistics Accounting for Linkage Disequilibrium and Horizontal Pleiotropy