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

Area of Science:

• Genetics
• Epidemiology
• Bioinformatics

Background:

• Mendelian randomization (MR) uses genetic variants as instrumental variables (IVs) to infer causal relationships from genome-wide association studies (GWAS).
• Violations of MR assumptions, particularly correlated horizontal pleiotropy (CHP) due to unmeasured confounders like shared genes or pathways, can bias causal effect estimates.
• Identifying and addressing CHP is crucial for robust causal inference in genetic epidemiology.

Purpose of the Study:

• To introduce MR-CUE (MR with Correlated horizontal pleiotropy Unraveling shared Etiology and confounding), a novel method for Mendelian randomization.
• To estimate causal effects while simultaneously identifying IVs exhibiting CHP and accounting for associated estimation uncertainty.
• To investigate shared genetic etiology underlying exposures and outcomes by mapping cis-associated genes and enriched pathways of pleiotropic IVs.

Main Methods:

• MR-CUE employs a statistical framework to detect IVs associated with unmeasured confounders.
• It quantifies causal effects while adjusting for estimation uncertainty introduced by CHP.
• The method integrates gene mapping and pathway enrichment analyses for pleiotropic IVs to explore shared genetic underpinnings.

Main Results:

• Application of MR-CUE to interleukin 6 (IL-6) revealed causal effects on multiple traits/diseases.
• Several S100 genes were identified as potentially involved in the shared genetic etiology between IL-6 and the studied outcomes.
• The study assessed the impact of various exposures on type 2 diabetes risk across diverse ancestries (European and East Asian populations).

Conclusions:

• MR-CUE provides a robust approach to Mendelian randomization by addressing correlated horizontal pleiotropy.
• The method facilitates the discovery of shared genetic etiology, offering insights into complex disease mechanisms.
• MR-CUE enhances the reliability of causal inference from genetic data and has broad applicability in genetic epidemiology.