Skip to content
Get packs

Arboreto Gene Network Inference

Ranks candidate gene regulator-target links from expression data using GRNBoost2 or GENIE3.

Source Download skill
Repository
K-Dense-AI/scientific-agent-skills
Stars
55,780
Licence
MIT
Files
5 files

What it does

This skill teaches an agent to run Arboreto, a bioinformatics tool, on a table of gene expression measurements (rows are cells or samples, columns are genes). It produces a ranked list of which genes may regulate which other genes, based on predictive associations rather than proof of direct regulation. It also guides the agent through a pinned Python 3.11 environment, input checks, a bundled command-line wrapper, and stability checks across seeds.

Great for

  • Bioinformatics researchers building candidate gene regulatory networks from bulk or single-cell expression data
  • Computational biologists who want a reproducible, pinned Arboreto setup on Python 3.11
  • Analysts comparing GRNBoost2 and GENIE3 results across runs or conditions

Skip it if

  • Claiming causal or direct regulation from output alone, since scores are observational associations (needs motif, binding or perturbation evidence)
  • Users expecting the latest PyPI install to work out of the box, since the skill says PyPI release 0.1.6 needs specific pinned dependencies and newer source fixes are not in the wheel
  • Non-biology datasets or people without expression-matrix experience, as the inputs and interpretation require domain knowledge

Try asking

  • “Infer a candidate gene regulatory network from my single-cell expression TSV using these TF names: tfs.txt”
  • “Run GRNBoost2 with seed 777 on a small subset first, then scale to 4 workers”
  • “Compare GRNBoost2 and GENIE3 results on the same expression matrix and summarize overlap”
  • “Check stability of the top TF-target edges across several seeds and resampled observations”
Step-by-step guide: how to use Arboreto Gene Network InferenceInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 5 original files, the MIT licence and an attribution note.

    arboreto.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip arboreto.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip arboreto.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Always run a small pilot subset in the pinned environment before scaling worker counts, and keep the __main__ guard in scripts.
  • Provide a TF list via --tf-file; without it, all genes are treated as candidate regulators.
  • Archive seeds, package versions, filters and identifier lists, and count edge retention across all runs rather than averaging only present rows.

What's inside

  • references/algorithms.md
  • references/basic_inference.md
  • references/distributed_computing.md
  • scripts/basic_grn_inference.py
  • SKILL.md

Redistributed unmodified from K-Dense-AI/scientific-agent-skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.