from alterlab-academic-skills18
Bayesian modeling workflow using PyMC (v5+): build hierarchical models, run MCMC (NUTS) or variational inference, diagnose sampling, and compare models with LOO
Provides a complete Bayesian modeling workflow using PyMC (v5+). Covers data prep, model building (linear, logistic, hierarchical), prior predictive checks, MCMC (NUTS) inference, variational inference, diagnostics, posterior predictive checks, and model comparison (LOO/WAIC). Includes templates and diagnostic scripts to produce reproducible analyses.
scripts/model_diagnostics.py, scripts/model_comparison.py and other helper scripts (has_scripts=true)references/ with distribution guides, sampling/inference notes, and workflow cookbooks (has_references=true)assets/ (linear_regression_template.py, hierarchical_model_template.py), and diagnostic/reporting utilitiesUseful for coding assistants and data-science-focused agents that can run Python and inspect repo assets (Codex, Claude Code, Cursor).
This skill has not been reviewed by our automated audit pipeline yet.
AlterLab Deep Research
A 13-agent, PRISMA-capable research pipeline that runs scoping, systematic literature search, verification, synthesis, meta-analysis, and APA‑style reporting fo
ClinVar Database (AlterLab)
Search and interpret NCBI ClinVar variant data, access via E-utilities or FTP, annotate VCFs, and incorporate review-status and evidence best practices for geno
