MendelianRandomization

MendelianRandomization performs Mendelian randomization analyses on summarized genetic data in R to estimate causal relationships between exposures and outcomes using instrumental variable methods.

Key Features:

• Core Methods: Implements instrumental variable methods including inverse-variance weighted (IVW), MR-Egger regression, and weighted median approaches for deriving causal estimates from multiple genetic variants.
• Advanced Options: Provides robust regression techniques, fixed-effects and random-effects models, and an option to penalize weights for genetic variants with heterogeneous causal estimates.
• Correlation Handling: Includes extensions to account for correlations among genetic variants.
• Graphical Capabilities: Offers graphical commands to visualize summarized data and compare causal estimates derived from different methods in a single plot.
• Data Integration: Facilitates integration with PhenoScanner to incorporate genotype-phenotype association data into analyses.

Scientific Applications:

• Epidemiology: Infers causal effects of exposures such as lifestyle or environmental factors on outcomes including diseases and quantitative traits using genetic instruments.
• Genetics: Supports genetic epidemiology studies that use summarized genetic associations to disentangle causal relationships between variants, exposures, and phenotypes.
• Public Health: Provides causal evidence to inform risk factor prioritization and potential intervention targets in population health research.

Methodology:

Uses instrumental variable analysis with genetic variants as instruments and implements inverse-variance weighted (IVW), MR-Egger regression, weighted median, robust regression, fixed-effects and random-effects models, penalized weights for heterogeneous variants, extensions for correlated variants, graphical comparison of causal estimates, and integration with PhenoScanner genotype-phenotype associations.