Arboreto

This is a Python library from Aerts Lab that infers gene regulatory networks from RNA-seq data using tree-based ensemble methods. You get two algorithms: GRNBoost2 (faster, gradient boosting) and GENIE3 (classic random forest). Feed it an expression matrix, get back transcription factor to target gene relationships with importance scores. It uses Dask for parallelization, so you can scale from local cores to clusters for large single-cell datasets. The library is mature but not actively maintained (last PyPI release 2021), and it's the GRN inference engine behind pySCENIC. Main gotcha is you need the if __name__ == '__main__' guard everywhere because of Dask's multiprocessing. Set a seed for reproducibility and filter outputs by importance thresholds.

Install to Claude Code

npx -y skills add k-dense-ai/scientific-agent-skills --skill arboreto --agent claude-code