A specialized AI agent skill for generating comprehensive, publication-ready two-sample Mendelian randomization research designs from a research direction.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install two-sample-mr-research-planner-2
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install two-sample-mr-research-planner-2 using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Two-Sample MR Research Planner?
The Two-Sample MR Research Planner is a sophisticated technical framework designed to automate the complex planning phase of genetic epidemiology. By leveraging Openclaw Skills, researchers can transform a simple exposure-outcome hypothesis into a detailed, multi-tier study protocol. It streamlines the identification of GWAS data sources, instrument selection criteria, and the statistical methodology required to test causal relationships between traits.
This skill is indispensable for bioinformaticians and medical researchers looking to build rigorous causal inference pipelines using summary-level genetic data. Using Openclaw Skills like this one ensures that all methodological pitfalls, from weak instrument bias to horizontal pleiotropy, are addressed in the initial design phase. It provides a technical roadmap that bridges the gap between raw data and high-impact scientific publications.
Two-Sample MR Research Planner Use Cases
- Designing a publishable Mendelian randomization paper from scratch based on a specific biomarker.
- Screening a panel of metabolic exposures to identify causal drivers of chronic disease.
- Planning bidirectional MR studies to investigate reciprocal causal relationships and rule out reverse causation.
- Developing advanced protocols involving multivariable MR (MVMR) to adjust for pleiotropic confounding.
- Generating rapid-turnaround Lite configurations for pilot data verification within the Openclaw Skills ecosystem.
How Two-Sample MR Research Planner Works
- Input Analysis: The skill analyzes the user provided research direction to identify exposure classes and outcome phenotypes.
- Workload Tiering: It automatically generates four distinct configurations (Lite, Standard, Advanced, and Publication+) tailored to different resource constraints.
- Primary Recommendation: The agent selects the optimal configuration based on exposure-class IV count benchmarks and data availability.
- Technical Workflow Mapping: It defines a step-by-step execution path including instrument extraction, LD clumping, and harmonization.
- Sensitivity Protocol: The skill outlines a robust validation strategy using MR-Egger, weighted median, and leave-one-out analyses.
- Publication Strategy: It generates a figure plan and deliverable list intended for peer-reviewed journal submission.
Two-Sample MR Research Planner Setup
To implement the research designs generated by this skill, you must have an R environment configured with the appropriate libraries. This tool integrates perfectly with other Openclaw Skills to manage the transition from planning to execution.
# Install the core R packages for Mendelian Randomization
install.packages("TwoSampleMR")
install.packages("MRPRESSO")
Ensure you have access to the IEU Open GWAS API to resolve the example IDs provided in the study plans.
Two-Sample MR Research Planner Data Schema & Taxonomy
The skill organizes research designs using a structured taxonomy of GWAS metadata and statistical thresholds. This organization is a core feature of research-oriented Openclaw Skills.
| Data Layer | Key Metadata Fields |
|---|---|
| Exposure Data | GWAS ID, P-value threshold (5e-08), LD clumping (r2 < 0.001), F-statistic |
| Outcome Data | SNP intersection, proxy matching, palindromic SNP handling |
| Core Estimates | IVW (Primary), MR-Egger, Weighted Median, Simple/Weighted Mode |
| Sensitivity | Cochran's Q, I2, MR-PRESSO global test, intercept p-value |
Two-Sample MR Research Planner Advanced Features
- Multi-tier configuration generation providing Lite to Publication+ workload options.
- Built-in fault tolerance for handling weak instrument risks (F-statistic < 10).
- Specialized workflows for molecular trait MR (proteins, metabolites) including mandatory colocalization steps.
- Detailed risk review section that anticipates reviewer criticisms and proposes revision strategies.
- Seamless integration with the broader library of Openclaw Skills for end-to-end bioinformatics automation.
SKILL.md
Loading
Related Openclaw Skills
OpenClaw Local Security Audit
A comprehensive security auditing tool designed to identify, verify, and resolve vulnerabilities in local OpenClaw deployments and sandbox environments.
LobsterDomains
A high-performance AI agent skill for searching and registering domain names using cryptocurrency payments across multiple blockchain networks.
Chemical Storage Sorter
A specialized tool for classifying laboratory chemicals and organizing them into safe storage groups based on compatibility and hazard regulations.
Tencent Cloud FaceID DetectLiveFaceAccurate
A high-precision liveness detection skill that identifies whether a face in a static image is a real person or a fraudulent attempt like a screen capture or 3D mask.
Clean Unused Dependencies
A utility to streamline project maintenance by detecting and removing redundant npm dependencies.
Safety Executor
A specialized safety framework for AI agents to execute system-level commands while preventing destructive operations through hard blocks and user confirmations.
Featured*
