Mendelian Randomization Tools

Configuration File: mendelian_randomization_tools.json Tool Type: Local Tools Count: 1

This page contains all tools defined in the mendelian_randomization_tools.json configuration file.

Available Tools

MR_estimate (Type: MendelianRandomizationTool)

Two-sample Mendelian randomization causal estimate from harmonized instrument summary statistics …

MR_estimate tool specification

Tool Information:

  • Name: MR_estimate

  • Type: MendelianRandomizationTool

  • Description: Two-sample Mendelian randomization causal estimate from harmonized instrument summary statistics (SNP-exposure + SNP-outcome betas and SEs, e.g. from OpenGWAS_get_mr_instruments). Returns IVW (primary), MR-Egger (with directional-pleiotropy intercept), weighted-median estimates, and Cochran’s Q / I^2 heterogeneity. Pure-compute (no R, no API key); reimplements TwoSampleMR’s core estimators.

Parameters:

  • beta_exposure (array) (required) Per-instrument SNP-exposure effect sizes (harmonized to the same effect allele; non-zero).

  • se_exposure (array) (required) Standard errors of beta_exposure (positive).

  • beta_outcome (array) (required) Per-instrument SNP-outcome effect sizes (same SNP order / effect allele).

  • se_outcome (array) (required) Standard errors of beta_outcome (positive).

Example Usage:

query = {
    "name": "MR_estimate",
    "arguments": {
        "beta_exposure": ["item1", "item2"],
        "se_exposure": ["item1", "item2"],
        "beta_outcome": ["item1", "item2"],
        "se_outcome": ["item1", "item2"]
    }
}
result = tu.run(query)