Multi-trait analysis of rare-variant association summary statistics using MTAR

Abstract

Integrating association evidence across multiple traits can improve the power of gene discovery and reveal pleiotropy. Most multi-trait analysis methods focus on individual common variants in genome-wide association studies. Here, we introduce multi-trait analysis of rarevariant associations (MTAR), a framework for joint analysis of association summary statistics between multiple rare variants and different traits. MTAR achieves substantial power gain by leveraging the genome-wide genetic correlation measure to inform the degree of gene-level effect heterogeneity across traits. We apply MTAR to rare-variant summary statistics for three lipid traits in the Global Lipids Genetics Consortium. 99 genome-wide significant genes were identified in the single-trait-based tests, and MTAR increases this to 139. Among the 11 novel lipid-associated genes discovered by MTAR, 7 are replicated in an independent UK Biobank GWAS analysis. Our study demonstrates that MTAR is substantially more powerful than single-trait-based tests and highlights the value of MTAR for novel gene discovery.

Genetic mapping of developmental trajectories for complex traits and diseases

“…5 C). Top candidate genes identified for the liver development – HDL link include APOH [47] , ITIH3 [48] , and APOE [49] ( Fig. 5 D).…”

Breeding for Climate Change Resilience: A Case Study of Loblolly Pine (Pinus taeda L.) in North America

“…Many of the SNPs associated with phenotypic or environmental variation also show low minor-allele frequency in the population, suggesting that many such loci would have to be genotyped in many individual trees in order to have enough power to model phenotypic variation for breeding purposes. Multivariate methods that analyze multiple SNP alleles (sometimes grouped by functional categories such as the genes in which those SNPs occur) and multiple phenotypes in parallel have shown improved power to detect associations of genotypes with environmental variables (Rellstab et al, 2015;Forester et al, 2018) and with phenotypic variation (Kaakinen et al, 2017;Luo et al, 2020). Genomic selection is another approach to parallel analysis of many SNP loci with phenotypic information, and considerable interest has been shown in applying this method to forest tree breeding (reviewed by Lebedev et al, 2020).…”

Nontrivial Replication of Loci Detected by Multi-Trait Methods

“…However, there are at least two disadvantages of such "univariate" replication: firstly, when the number of tested traits is large, multiple testing arises when determining the replication significance threshold; secondly, univariate replication does not account for phenotypic correlations between the tested traits, which generates conservative significance threshold after correction for multiple testing. Another straightforward way for replication is to directly perform the multi-trait test in a replication sample and see whether the overall association (omnibus p-value) is significant (Karnes et al, 2017;Liang et al, 2017;Luo et al, 2020). Although this strategy provides a unified test statistic, it does not reveal whether the effects that locus exhibits on traits in the discovery are the same (or similar) as those observed in replication.…”