bio-causal-genomics-pleiotropy-detection

name	bio-causal-genomics-pleiotropy-detection
description	Detect and correct for horizontal pleiotropy in Mendelian randomization analyses using MR-PRESSO for outlier removal, MR-Egger regression for directional pleiotropy, and Steiger filtering for variant directionality. Use when validating MR results, detecting pleiotropic instrum...
license	MIT
author	AIPOCH

Source: https://github.com/aipoch/medical-research-skills

Version Compatibility

Reference examples tested with: MR-PRESSO 1.0+, TwoSampleMR 0.5+

Before using code patterns, verify installed versions match. If versions differ:

• R: packageVersion("<pkg>") then ?function_name to verify parameters

If code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying.

Pleiotropy Detection

"Check my MR results for pleiotropic bias" → Detect and correct for horizontal pleiotropy using outlier removal (MR-PRESSO), directional pleiotropy testing (MR-Egger intercept), and variant directionality filtering (Steiger) to validate causal inference results.

• R: MRPRESSO::mr_presso() for global and distortion tests
• R: TwoSampleMR::mr_egger_regression() for Egger intercept test

Overview

Horizontal pleiotropy violates the exclusion restriction assumption of MR: instruments affect the outcome through pathways other than the exposure. Detecting and correcting for pleiotropy is essential for valid causal inference.

Types of pleiotropy:

• Vertical (mediated): Instrument -> exposure -> outcome (valid, not a problem)
• Horizontal (direct): Instrument -> outcome bypassing exposure (violates MR assumptions)
• Balanced: Pleiotropic effects cancel out (IVW still valid, Egger intercept ~0)
• Directional: Pleiotropic effects are systematic (biases IVW, Egger detects this)

MR-PRESSO

Goal: Detect and remove pleiotropic outlier instruments from an MR analysis.

Approach: Run MR-PRESSO to test for global pleiotropy, identify individual outlier SNPs, test whether their removal changes the causal estimate (distortion test), and obtain a corrected estimate.

# install.packages('remotes')
# remotes::install_github('rondolab/MR-PRESSO')
library(MRPRESSO)

# Input: harmonized data from TwoSampleMR
# Columns needed: beta.exposure, beta.outcome, se.exposure, se.outcome
presso_input <- data.frame(
  bx = dat$beta.exposure,
  by = dat$beta.outcome,
  bxse = dat$se.exposure,
  byse = dat$se.outcome
)

# --- Run MR-PRESSO ---
# NbDistribution: Number of simulations for null distribution (minimum 1000)
# SignifThreshold: P-value threshold for outlier detection (0.05 standard)
presso_result <- mr_presso(
  BetaOutcome = 'by', BetaExposure = 'bx',
  SdOutcome = 'byse', SdExposure = 'bxse',
  OUTLIERtest = TRUE, DISTORTIONtest = TRUE,
  data = presso_input,
  NbDistribution = 5000,
  SignifThreshold = 





global_p  presso_result`MR-PRESSO results``Global Test`Pvalue
cat global_p 



outliers  presso_result`MR-PRESSO results``Outlier Test`
cat
printoutliers


outlier_indices  whichoutliersPvalue  
cat outlier_indices 




distortion_p  presso_result`MR-PRESSO results``Distortion Test`Pvalue
cat distortion_p 



main_results  presso_result`Main MR results`
cat main_results`Causal Estimate` 
cat main_results`Causal Estimate` 

MR-Egger Diagnostics

Goal: Test for directional pleiotropy and obtain a pleiotropy-adjusted causal estimate.

Approach: Fit MR-Egger regression where the intercept estimates average pleiotropic bias, and check I-squared for instrument strength under the NOME assumption.

library(TwoSampleMR)

# MR-Egger regression allows for a non-zero intercept
# The intercept estimates the average pleiotropic effect
egger <- mr_egger_regression(dat$beta.exposure, dat$beta.outcome,
                              dat$se.exposure, dat$se.outcome)

# --- Egger intercept ---
# Significant intercept (p < 0.05): Directional pleiotropy present
# Non-significant: No evidence (but low power with < 10 SNPs)
cat('Egger intercept:', round(egger$b_i, 5), '\n')
cat('Intercept SE:', round(egger$se_i, 5), '\n')
cat('Intercept p-value:', format.pval(egger$pval_i), '\n')

# --- Egger slope ---
# Valid causal estimate EVEN with directional pleiotropy (InSIDE assumption)
cat eggerb  
cat eggerse  
cat format.pvaleggerpval 





isq  Isqdatbeta.exposure datse.exposure
cat isq  
 isq   cat

Steiger Filtering

Goal: Verify that instruments act in the correct causal direction (exposure -> outcome, not reverse).

Approach: Apply the Steiger test to each instrument, remove those explaining more outcome variance than exposure variance, and re-run MR on filtered instruments.

library(TwoSampleMR)

# Steiger test: Verify each instrument explains more variance in
# the exposure than the outcome. Instruments failing this test
# may act through a reverse causal pathway.

steiger <- steiger_filtering(dat)

# Keep only correctly oriented instruments
dat_steiger <- steiger[steiger$steiger_dir == TRUE, ]
cat('Instruments passing Steiger filter:', nrow(dat_steiger), 'of', nrow(steiger), '\n')

# Re-run MR with filtered instruments
results_steiger <- mr(dat_steiger)
print(results_steiger[, c('method', 'nsnp', 'b', 'se', 'pval')])

# Directionality test (aggregate)
direction <- directionality_testdat
cat directioncorrect_causal_direction 
cat format.pvaldirectionsteiger_pval 

Additional Sensitivity Methods

library(TwoSampleMR)

# --- Contamination mixture ---
# Assumes some instruments are valid, others are not
# Does not require majority valid assumption
mr_conmix <- mr(dat, method_list = 'mr_raps')

# --- MR-RAPS ---
# NOTE: MRAPS CRAN package was archived March 2025.
# Use the MendelianRandomization package instead, or install from GitHub:
# remotes::install_github('qingyuanzhao/mr.raps')
library(MendelianRandomization)

mr_input <- mr_input(
  bx = dat$beta.exposure, bxse = dat$se.exposure,
  by = dat$beta.outcome, byse = dat$se.outcome
)

raps_result <- mr_raps(mr_input)
cat('MR-RAPS estimate:', raps_result$Estimate, '\n')
cat('MR-RAPS p-value:', raps_result$Pvalue, '\n')

# --- Multivariable MR ---


exposure1  extract_instruments  
exposure2  extract_instruments  

Comprehensive Sensitivity Framework

Goal: Run a complete battery of MR sensitivity analyses to validate causal findings.

Approach: Apply IVW, MR-Egger, weighted median, weighted mode, heterogeneity, Egger intercept, leave-one-out, and MR-PRESSO in a single function and summarize results.

library(TwoSampleMR)
library(MRPRESSO)

run_sensitivity <- function(dat) {
  results <- list()

  # 1. IVW (primary)
  results$ivw <- mr(dat, method_list = 'mr_ivw')

  # 2. MR-Egger
  results$egger <- mr(dat, method_list = 'mr_egger_regression')

  # 3. Weighted median (robust to 50% invalid instruments)
  results$median <- mr(dat, method_list = 'mr_weighted_median')

  # 4. Weighted mode
  results$mode <- mr(dat, method_list = 'mr_weighted_mode')

  # 5. Heterogeneity
  results$het <- mr_heterogeneity(dat)

  # 6. Egger intercept
  results$pleio <- mr_pleiotropy_testdat

  
  resultsloo  mr_leaveoneoutdat

  
  presso_input  data.frame
    bx  datbeta.exposure by  datbeta.outcome
    bxse  datse.exposure byse  datse.outcome
  
  resultspresso  mr_presso
    BetaOutcome   BetaExposure  
    SdOutcome   SdExposure  
    OUTLIERtest   DISTORTIONtest  
    data  presso_input NbDistribution   SignifThreshold  
  

  results


summarize_sensitivity  sens 
  cat

  
  all_mr  rbindsensivw sensegger sensmedian sensmode
  cat
  printall_mr    

  
  cat
  cat senshetQ_pvalsenshetmethod   

  
  cat
  cat senspleioegger_intercept 
  cat senspleiopval 

  
  cat
      senspresso`MR-PRESSO results``Global Test`Pvalue 

  cat
  cat
  cat
  cat
  cat

STROBE-MR Reporting

When reporting MR analyses, follow STROBE-MR guidelines:

1. Report all MR methods tested (not just the most significant)
2. Report heterogeneity Q-statistic and p-value
3. Report Egger intercept with p-value
4. Report MR-PRESSO global test and number of outliers removed
5. Report F-statistics for instrument strength
6. Report Steiger directionality test
7. State whether results are consistent across sensitivity analyses
8. Acknowledge limitations of the MR assumptions

Related Skills

• mendelian-randomization - Primary MR analysis that pleiotropy tests validate
• fine-mapping - Identify causal variants at instrument loci
• population-genetics/association-testing - GWAS data for MR instruments

Input Validation

This skill accepts requests that match the documented purpose of bio-causal-genomics-pleiotropy-detection and include enough context to complete the workflow safely.

Do not continue the workflow when the request is out of scope, missing a critical input, or would require unsupported assumptions. Instead respond:

bio-causal-genomics-pleiotropy-detection only handles its documented workflow. Please provide the missing required inputs or switch to a more suitable skill.