Claude Skill MartBrowse skillsWhat is a Skill?
 
← Back to list

Gtars Genomic Interval Analysis

A guardrailed guide for using gtars across Python, Rust, and the CLI for BED set algebra, coverage, consensus, tokenization, and refget.

Data & AnalyticsAdvanced★ 33,030⑂ 3,248AI score 8/10Last updated: Aug 9, 2026

What it does

This skill teaches Claude to drive the gtars genomic-interval toolkit correctly and safely.

  • Overlaps & set algebra: count_overlaps, any_overlaps, find_overlaps, intersect_all, jaccard, setdiff, reduce, cluster, gaps against the real Python 0.9.2 API
  • Consensus & coverage: build consensus interval sets via genomic_distributions.consensus; clearly separates coverage metrics from signal-track (uniwig) generation
  • Tokenizers, fragments, refget: local universe-BED tokenizers plus rules for in-memory/local/remote RefgetStore
  • Enforced data contract: 0-based half-open coordinates, u32 bounds, recorded assembly accession + SHA-256, exact contig-name matching, explicit sort and strand policies
  • Six bundled local CLIs: BED validator, execution planner, tokenizer manifest, refget digest plan, coverage preflight, artifact inspector — no network, no output files
  • Safety gates: PyO3 wheels and Cargo build scripts are treated as code execution; downloads and cache writes require explicit approval

Who it's for

  • Bioinformaticians working daily with BED files (ATAC-seq/ChIP-seq peaks, promoters, enhancers)
  • Teams building interval-based ML pipelines (universes, tokenizers)
  • Anyone who wants to avoid coordinate-system bugs, contig-name mismatches, and train/test leakage
  • Groups needing patient/donor-level splits and reproducible provenance records

Usage examples

  1. Overlap summary: "How much of peaks.bed overlaps promoters.bed?" → validate coordinates and sorting, then generate count_overlaps / jaccard code
  2. Consensus set: merge many sample BEDs into a consensus universe fit on training data only, preventing validation/test locus leakage
  3. Tokenizer prep: skip unapproved Hugging Face downloads and build Tokenizer.from_bed from a reviewed local universe, recording a manifest and checksums

· · · Install guide · · ·

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/gtars folder from the GitHub repo K-Dense-AI/scientific-agent-skills into my ~/.claude/skills/gtars/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/K-Dense-AI/scientific-agent-skills.git && mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/gtars ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal and clone the repo: git clone https://github.com/K-Dense-AI/scientific-agent-skills.git
  2. Create the skills directory: mkdir -p ~/.claude/skills
  3. Copy the skill: cp -r scientific-agent-skills/skills/gtars ~/.claude/skills/
  4. Confirm ~/.claude/skills/gtars/ contains SKILL.md plus the references/ and scripts/ folders.
  5. Restart Claude Code, then try a prompt like "analyze overlaps between these two BED files".
  6. (Optional) To actually run gtars, create a Python 3.10+ virtual environment and install it, e.g. uv pip install "gtars==0.9.2". Verify the real current version and publisher on PyPI first — the pins listed in the skill may be out of date.
  7. (Optional) For the CLI, install a Rust toolchain and run cargo install gtars-cli --locked.
  8. For network-capable features (pretrained tokenizers, remote refget, BEDbase cache) the skill will ask for approval, so decide your allowed hosts and cache paths in advance.
View source on GitHub ↗License: MIT