Integrating genetic data with biological insight: A practical guide to cis-Mendelian randomization

Area of Science:

• Genetics
• Computational Biology
• Epidemiology

Background:

• Cis-Mendelian randomization (cis-MR) is increasingly used to infer causality.
• Abundant genetic data for molecular traits fuels cis-MR adoption.
• Guidance is needed for robust cis-MR study design and execution.

Purpose of the Study:

• To review and discuss key considerations for conducting robust cis-Mendelian randomization analyses.
• To provide informed guidance for researchers utilizing cis-MR.
• To improve the standard of planning, conducting, reviewing, and interpreting cis-MR studies.

Main Methods:

• Review of existing cis-Mendelian randomization methodologies.
• Discussion of critical factors including gene region selection, trait proxy choice, variant selection/validation, and sensitivity analyses.
• Comparison with genome-wide Mendelian randomization.

Main Results:

• Identified key considerations for robust cis-MR: gene region selection, trait proxy choice, variant validation, and sensitivity analyses.
• Highlighted the importance of biological insight and tailored analytical methods.
• Emphasized distinct sensitivity analyses required for cis-MR compared to genome-wide MR.

Conclusions:

• Implementing the discussed considerations will enhance the rigor of cis-Mendelian randomization studies.
• Tailoring methods and incorporating biological insight are crucial for valid cis-MR.
• This review aims to elevate the quality of research using cis-Mendelian randomization.