Dietary Intake Mendelian Randomization: Assessment and Development of Methods for Instrument Selection and Robust...

Area of Science:

• Genetics and Epidemiology
• Nutritional Science
• Biostatistics

Background:

• Mendelian randomization (MR) is used to infer causality between dietary intake and health outcomes.
• Current MR approaches often use all genome-wide significant (GWS) variants, potentially introducing bias from indirect effects.

Purpose of the Study:

• To assess the current literature on diet MR and identify common methodologies.
• To evaluate the efficacy of conventional MR techniques in detecting established diet-health relationships.
• To develop and test methods for refining genetic instruments (GIs) to better estimate direct dietary effects.

Main Methods:

• Literature review of diet MR studies identified via PubMed.
• Empirical testing of conventional two-sample MR with GWS variants on six positive control diet-health relationships.
• Development and application of GI refinement techniques: Phenome-wide association study (PheWAS)-based filtering and multivariable MR (MVMR).

Main Results:

• Conventional MR with GWS GIs showed limited success in detecting expected diet-health associations.
• GI filtering improved detection for diet-biomarker pairs but not diet-disease pairs.
• MR-CAUSE identified a diet-disease association (white vs. whole grain bread on CVD) and highlighted confounding by body composition traits.

Conclusions:

• The widespread use of MR for dietary exposures necessitates caution due to inconsistent findings and potential confounding.
• Thoughtful curation of genetic instruments is crucial for reducing bias and accurately assessing the causal role of diet in health.