MRLocus: Identifying causal genes mediating a trait through Bayesian estimation of allelic heterogeneity

Abstract

Expression quantitative trait loci (eQTL) studies are used to understand the regulatory function of non-coding genome-wide association study (GWAS) risk loci, but colocalization alone does not demonstrate a causal relationship of gene expression affecting a trait. Evidence for mediation, that perturbation of gene expression in a given tissue or developmental context will induce a change in the downstream GWAS trait, can be provided by two-sample Mendelian Randomization (MR). Here, we introduce a new statistical method, MRLocus, for Bayesian estimation of the gene-to-trait effect from eQTL and GWAS summary data for loci with evidence of allelic heterogeneity, that is, containing multiple causal variants. MRLocus makes use of a colocalization step applied to each nearly-LD-independent eQTL, followed by an MR analysis step across eQTLs. Additionally, our method involves estimation of the extent of allelic heterogeneity through a dispersion parameter, indicating variable mediation effects from each individual eQTL on the downstream trait. Our method is evaluated against other state-of-the-art methods for estimation of the gene-to-trait mediation effect, using an existing simulation framework. In simulation, MRLocus often has the highest accuracy among competing methods, and in each case provides more accurate estimation of uncertainty as assessed through interval coverage. MRLocus is then applied to five candidate causal genes for mediation of particular GWAS traits, where gene-to-trait effects are concordant with those previously reported. We find that MRLocus’s estimation of the causal effect across eQTLs within a locus provides useful information for determining how perturbation of gene expression or individual regulatory elements will affect downstream traits. The MRLocus method is implemented as an R package available at https://mikelove.github.io/mrlocus.

Combining evidence from Mendelian randomization and colocalization: Review and comparison of approaches

“… 94 Another related method is MRLocus, which combines a colocalization step and a Mendelian randomization slope fitting step in a Bayesian hierarchical model, allowing for multiple causal variants and allelic heterogeneity. 95 …”

Probabilistic integration of transcriptome-wide association studies and colocalization analysis prioritizes molecular pathways of complex traits

“…The first observation frames PCG implication as a model selection/hypothesis testing problem and distinguishes it from a causal inference problem aiming to estimate γ unbiasedly (e.g., MR-Locus [15]). It is sufficient to establish γ ≠ 0 by showing the correlation between the genetically predicted gene expression levels, , and Y is non-zero under the necessary assumptions of IV analysis [24, 14, 6].…”

“…The extent of the existing knowledge base can limit these methods, and they cannot quantify the uncertainty of the PCG implications. With the increasing availability of genome-scale molecular phenotyping, a new class of emerging computational methods has been developed to establish a relationship between genetic variants, molecular phenotypes, and complex traits by integrating GWAS and molecular quantitative trait loci (QTL) data [6, 7, 8, 9, 10, 11, 12, 13, 14, 15]. These methods have shown promise in not only identifying PCGs but also implicating relevant molecular mechanisms.…”

Bench Research Informed by GWAS Results

“…For causality estimation, it is possible to employ multiple LD-independent loci within the vicinity of a gene, which could be used to distinguish mediation and pleiotropic effects, since in a true causality situation, eQTL and GWAS effects should be correlated in multiple loci. This approach is realised in the MRLocus method [238].…”