deeptools — Claude Code skill: NGS analysis toolkit
A skill that turns sequencing data into coverage tracks, QC metrics and heatmaps around TSS and peaks.
npx skills add K-Dense-AI/scientific-agent-skills --skill deeptools -a claude-codeWhat deeptools does
The skill hands your agent the deepTools command set for high-throughput sequencing work. It converts BAM alignments into normalized coverage tracks, computes quality metrics, compares samples and draws heatmaps with profiles around genomic features. Reach for it when you need to process ChIP-seq, RNA-seq or ATAC-seq and produce figures for a paper or a report.
Two scripts ship with it: scripts/validate_files.py and scripts/workflow_generator.py. The first checks local BAM, bigWig and BED files — structure, coordinate order, headers, every BED row. The second builds a runnable script for a standard scenario such as chipseq_qc, filling in your file names. Reference notes cover normalization methods, effective genome sizes and per-tool options.
You need Python 3.12+ and deepTools 4.0.0 with pysam and pyBigWig; samtools handles sorting and indexing. Five commands were rewritten in Rust for 4.0.0: bamCoverage, bamCompare and multiBamSummary no longer take --ignoreDuplicates, and --exactScaling is gone. Pipelines still carrying those flags break at argument parsing, so check your commands before a cluster run.
What it helps with
- convert BAM alignments into normalized bigWig tracks for a genome browser
- check ChIP-seq quality with fingerprints and replicate correlation
- plot a heatmap of signal around TSS or peak regions
- compare treatment against control as a log2 ratio
- generate a runnable script for a standard workflow with your files
Example prompts for the agent
- Validate
sample1.bamandregions.bedwith the validation script - Generate the
chipseq_qcworkflow forChIP1.bamandChIP2.bam - Compute correlation between replicates and show me the result
How to install deeptools
1Skills CLI (Vercel)
npx skills add K-Dense-AI/scientific-agent-skills --skill deeptools -a claude-code2Manually: put the files in place
~/.claude/skills/deeptools/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/deeptools/
1Skills CLI (Vercel)
npx skills add K-Dense-AI/scientific-agent-skills --skill deeptools -a codex2Manually: put the files in place
.agents/skills/deeptools/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add K-Dense-AI/scientific-agent-skills --skill deeptools -a cursor2Manually: put the files in place
.cursor/skills/deeptools/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install K-Dense-AI/scientific-agent-skills deeptools2Manually: put the files in place
.github/skills/deeptools/SKILL.md1Skills CLI (Vercel)
npx skills add K-Dense-AI/scientific-agent-skills --skill deeptools -a gemini-cli2Platform CLI
gemini skills install https://github.com/K-Dense-AI/scientific-agent-skills --path skills/deeptoolsManually: put the files in place
.devin/skills/deeptools/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add K-Dense-AI/scientific-agent-skills --skill deeptools -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about deeptools
How do I install deeptools in Claude Code?
Run npx skills add K-Dense-AI/scientific-agent-skills --skill deeptools -a claude-code. The skill lands in ~/.claude/skills/deeptools/SKILL.md.
What does deeptools do?
It processes NGS data: converts BAM to bigWig, computes quality metrics, compares samples and draws heatmaps with profiles around genomic features.
What does the skill require?
Python 3.12+ and deepTools 4.0.0 with pysam and pyBigWig, plus samtools for sorting and indexing. Network access is only needed for installation or fetching public input data.
