Group Identification and Variable Selection in Multivariable Mendelian Randomization with Highly-Correlated Exposures

TL;DR

This paper introduces MVMR-PACS, a novel method for identifying causal groups among highly correlated risk factors in multivariable Mendelian Randomization, improving estimation accuracy and interpretability in high-dimensional genetic data.

Contribution

MVMR-PACS is the first method to identify signal-groups and provide valid post-selection inference in high-dimensional, highly correlated MVMR settings.

Findings

MVMR-PACS outperforms existing methods in simulations.

It identifies meaningful causal groups in lipoprotein traits.

Provides robust and interpretable causal effect estimates.

Abstract

Multivariable Mendelian Randomization (MVMR) estimates the direct causal effects of multiple risk factors on an outcome using genetic variants as instruments. The growing availability of summary-level genetic data has created opportunities to apply MVMR in high-dimensional settings with many strongly correlated candidate risk factors. However, existing methods face three major limitations: weak instrument bias, limited interpretability, and the absence of valid post-selection inference. Here we introduce MVMR-PACS, a method that identifies signal-groups -- sets of causal risk factors with high genetic correlation or indistinguishable causal effects -- and estimates the direct effect of each group. MVMR-PACS minimizes a debiased objective function that reduces weak instrument bias while yielding interpretable estimates with theoretical guarantees for variable selection. We adapt a data-thinning strategy to summary-data MVMR to enable valid post-selection inference. In simulations, MVMR-PACS outperforms existing approaches in both estimation accuracy and variable selection. When applied to 27 lipoprotein subfraction traits and coronary artery disease risk, MVMR-PACS identifies biologically meaningful and robust signal-groups with interpretable direct causal effects.