Efficient inference for genetic association studies with multiple outcomes

Abstract

SUMMARY Combined inference for heterogeneous high-dimensional data is critical in modern biology, where clinical and various kinds of molecular data may be available from a single study. Classical genetic association studies regress a single clinical outcome on many genetic variants one by one, but there is an increasing demand for joint analysis of many molecular outcomes and genetic variants in order to unravel functional interactions. Unfortunately, most existing approaches to joint modeling are either too simplistic to be powerful or are impracticable for computational reasons. Inspired by Richardson and others (2010, Bayesian Statistics 9), we consider a sparse multivariate regression model that allows simultaneous selection of predictors and associated responses. As Markov chain Monte Carlo (MCMC) inference on such models can be prohibitively slow when the number of genetic variants exceeds a few thousand, we propose a variational inference approach which produces posterior information very close to that of MCMC inference, at a much reduced computational cost. Extensive numerical experiments show that our approach outperforms popular variable selection methods and tailored Bayesian procedures, dealing within hours with problems involving hundreds of thousands of genetic variants and tens to hundreds of clinical or molecular outcomes.

EPISPOT: an epigenome-driven approach for detecting and interpreting hotspots in molecular QTL studies

“…Indeed, widely-used marginal screening approaches [1,2] suffer from a large multiplicity burden and tend to lack of statistical power as they do not exploit the regulation patterns shared by the molecular entities, whereas joint modelling approaches [3,4] are often limited by the computational burden implied by the exploration of high-dimensional spaces of candidate variants and traits. To manage this tension between scalable inference and comprehensive joint modelling, we recently proposed a variational inference approach, called ATLASQTL [5], which explicitly borrows information across thousands of molecular traits controlled by shared pathways, and offers a robust fully Bayesian parametrisation of hotspots; its increased sensitivity and that of earlier related models have been demonstrated in different molecular QTL studies [4][5][6][7].…”

A fully joint Bayesian quantitative trait locus mapping of human protein abundance in plasma

“…The high replication rates and the novel discoveries uncovered by LOCUS are largely attributable to its flexible hierarchical sparse regression model which exploits shared association patterns across all SNPs and proteomic levels (Figures 1A-C), as extensively shown in previous numerical experiments. 11 Here, we provide additional evidence for our specific study and data by comparing LOCUS with the univariate method GEMMA in two ways. First, we evaluate variable selection performance in two simulation studies and, second, we confront the hits of LOCUS real data analysis to those found by re-analysing the Ottawa and…”

Bayesian sparse multiple regression for simultaneous rank reduction and variable selection

“…Table 2 reports sensitivity and specificity of this procedure and ours, averaged over 50 replicates. While the two methods performed almost identically in the relatively low dimensional setting (p, q) = (200, 30), BSML consistently outperformed Ruffieux et al (2017) when the dimension was higher. Table 1: Estimation and predictive performance of the proposed method (BSML) versus SPLS across different simulation settings.…”

“…To accommodate the non-diagonal error covariance, we placed a inverse-Wishart(q, I q ) prior on Σ. An associate editor pointed out the recent article (Ruffieux et al, 2017) which used spike-slab priors on the coefficients in a multiple response regression setting. They implemented a variational algorithm to posterior inclusion probabilities of each covariate, which is available from the R package locus.…”