tooluniverse-meta-analysis
Combine quantitative results from two or more studies into a pooled estimate and confidence interval, with built-in heterogeneity assessment (I², Q, τ²). The skill converts raw reported values—odds ratios, risk ratios, hazard ratios, means, proportions, correlations—into the standardized (effect, standard error) format pooling requires, then applies fixed- or random-effects models and generates a forest plot to visualize agreement across studies.
tooluniverse-meta-analysis pools effect sizes from multiple studies into a single estimate with confidence interval and heterogeneity metrics.
AI-generated summary based on this skill's SKILL.md
Install
mims-harvard/ToolUniverse/tooluniverse-meta-analysis · repository language: Python
git clone https://github.com/mims-harvard/ToolUniverse
cp -r ToolUniverse/skills/tooluniverse-meta-analysis ~/.claude/skills/tooluniverse-meta-analysisFrequently asked questions
AI-generated answers based on this skill's SKILL.md and metadata
What does tooluniverse-meta-analysis do?
tooluniverse-meta-analysis combines quantitative results from two or more studies into a pooled estimate and confidence interval, with built-in heterogeneity assessment (I², Q, τ²). It converts raw reported values—odds ratios, risk ratios, hazard ratios, means, proportions, correlations—into standardized (effect, standard error) format, then applies fixed- or random-effects models and generates a forest plot to visualize agreement across studies.
Can tooluniverse-meta-analysis pool effect sizes from multiple studies?
Yes. tooluniverse-meta-analysis is designed to pool effect sizes from multiple studies into a single estimate with confidence interval. It handles conversion of diverse study result formats—OR, RR, HR, means, proportions, and correlations—into the standardized effect and standard error format required for pooling, then synthesizes them using either fixed- or random-effects models depending on heterogeneity.
How does tooluniverse-meta-analysis assess heterogeneity?
tooluniverse-meta-analysis assesses heterogeneity using I², Q statistic, and τ² (tau-squared), helping you decide between fixed- and random-effects models. These metrics quantify the proportion of variation due to true study differences rather than sampling error, guiding model selection for accurate pooled estimates.
What formats can tooluniverse-meta-analysis convert to effect+SE?
tooluniverse-meta-analysis converts odds ratios, risk ratios, hazard ratios, means, proportions, and correlations (including Hedges g and mean differences) to standardized effect and standard error format. It handles log transformation of ratio measures and confidence interval-to-SE conversion, preparing all study results for pooling.
Does tooluniverse-meta-analysis generate forest plots?
Yes. tooluniverse-meta-analysis generates forest plots to visualize pooled estimates and individual study results, making it easy to see agreement or disagreement across studies. The plot displays heterogeneity visually alongside the synthesized systematic review evidence and overall pooled effect.
Is tooluniverse-meta-analysis suitable for multi-cohort GWAS?
tooluniverse-meta-analysis can synthesize multi-cohort association results by pooling effect sizes and standard errors across replicated experiments or cohorts. It applies random- or fixed-effects models to aggregate study results into a single estimate, making it applicable to large-scale evidence synthesis including GWAS meta-analysis.
SKILL.md
rendered from the published skill — quoted content, verbatim
Meta-Analysis & Evidence Synthesis
Pool quantitative results from multiple studies into one estimate, and judge how consistent the studies are. This is the statistical half of a systematic review (the literature-collection half is tooluniverse-literature-deep-research).
When to use this
- You have effect sizes from ≥2 studies/cohorts and want a single pooled estimate + CI.
- Synthesizing a systematic review, replicated experiments, multi-cohort GWAS, or multi-dataset
(truncated - see the full file via the links below)
Read as markdown · JSON record · Browse the source repository
File tree — 3 files
skills/tooluniverse-meta-analysis/SKILL.md
skills/tooluniverse-meta-analysis/scripts/meta_analysis.py
skills/tooluniverse-meta-analysis/test_meta_analysis.py