Does it include tools for detecting horizontal pleiotropy?

Absolutely. It covers the MR-Egger intercept test, Cochran’s Q test for heterogeneity, and MR-PRESSO for identifying and correcting for outlier-led pleiotropy.

How does this skill handle weak instrument bias?

The skill provides automated workflows for calculating F-statistics and performing clumping to ensure instruments meet the standard threshold of F > 10.

Can I perform Multivariable Mendelian Randomization (MVMR) with this?

Yes, it includes specific implementation patterns for MVMR to analyze multiple exposures simultaneously using both TwoSampleMR and the MendelianRandomization packages.

Which R packages does this skill support for MR analysis?

It supports industry standards including TwoSampleMR, MendelianRandomization, MRPRESSO, RadialMR, and specialized methods like CAUSE and MR-RAPS.

Mendelian Randomization in R

Name: Mendelian Randomization in R
Author: choxos

bychoxos

0•

Data Science & ML

Performs robust causal inference using genetic variants through a comprehensive suite of Mendelian Randomization (MR) methods and sensitivity analyses.

This skill equips Claude with specialized knowledge for conducting Mendelian Randomization (MR) studies within the R ecosystem. It provides standardized implementation patterns for instrument selection using GWAS databases, primary MR methods like IVW and MR-Egger, and rigorous sensitivity testing including pleiotropy assessment and MR-PRESSO. Whether performing two-sample MR, multivariable MR, or advanced causal analysis using methods like CAUSE, this skill ensures adherence to best practices and STROBE-MR reporting guidelines for high-quality epidemiological research.

Key Features

01Support for TwoSampleMR and MendelianRandomization package workflows.

02Advanced sensitivity analyses including MR-PRESSO outlier detection and Steiger filtering.

03Complex multivariable and bidirectional MR patterns for investigating interrelated exposures.

040 GitHub stars

05Publication-ready visualizations including scatter, forest, funnel, and radial plots.

06Automated instrument selection with LD clumping and F-statistic calculation for instrument strength.

Use Cases

01Validating observational epidemiological findings against potential confounding factors using genetic instruments.

02Determining causal relationships between clinical biomarkers and disease outcomes using GWAS summary statistics.

03Identifying pleiotropic effects and influential outliers in genetic data to improve the robustness of causal estimates.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add choxos/tidyrmodelling mendelian-randomization

For use in Claude.ai and ChatGPT

Download Skill