pku-embl/metagenomics-skills/checkm2
Claude Code and agent skill by pku-embl/metagenomics-skills
- community
Checking that the SKILL.md file is still in the repository.
ML-based MAG completeness and contamination estimation. Use for installation, parameter choice, execution, output interpretation, or troubleshooting.
Install
npx vskill@latest install pku-embl/metagenomics-skills/checkm2SKILL.md
CheckM2
Upstream: chklovski/CheckM2
Citation
Chklovski, A., Parks, D. H., Woodcroft, B. J. & Tyson, G. W. CheckM2: a rapid, scalable and accurate tool for assessing microbial genome quality using machine learning. Nat. Methods 20, 1203–1212 (2023). https://doi.org/10.1038/s41592-023-01940-w
See also docs/references.md.
Analytical thinking
CheckM1-style tools predict quality from lineage-specific single-copy markers. That fails or becomes overconfident when lineages lack robust marker sets (novel phyla/classes, CPR, etc.). CheckM2 uses gradient-boosting ML on broader genomic features to estimate completeness/contamination without explicit lineage marker sets — the paper emphasizes improved accuracy on taxonomically novel MAGs.
Still:
- Values are filters and reporting metrics, not wet-lab truth.
- Do not silently mix CheckM1 and CheckM2 scores across a study.
- Pair with
guncfor chimeric / clade-separated contamination CheckM-style
metrics can miss.
- BASALT refinement entry gates (e.g. comp≥35 / cont≤20) often differ from
publication filters (e.g. ≥90 / ≤5) — state both (mag-qc).
How to run
Key parameters
| Item | Notes | |
|---|---|---|
| Database | Pin CheckM2 DB release with software version | |
| Input | Directory of bin FASTAs | |
| Downstream | Feed filtered bins to drep / gtdbtk |
Validation, recovery, and provenance
Require quality_report.tsv, reconcile every input FASTA to exactly one successful or explicitly failed row, and reject duplicate genome IDs. Inspect completeness, contamination, model/translation-table warnings, genome size and coding density; compare discordant or novel genomes with $gunc, taxonomy, and marker evidence. Apply a predeclared quality policy and retain unrounded values.
Do not merge partial results from different database/software releases. After failure, use a clean output directory or document supported resume behavior. Record CheckM2 version, database release/path/checksum, command, input/output checksums, threads, failed genomes, and threshold decisions. Scores are predictions, not measured percentages.
Related skills
mag-qc · gunc · metawrap · basalt · drep · busco
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/checkm2/SKILL.md
Popularity
- Repository stars
- 7
- 7-day trend score
- 10
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/pku-embl/metagenomics-skills/checkm2)