菜单
Detect chimerism/contamination in prokaryotic genomes/MAGs. Use this skill when working with gunc: installing, choosing parameters, running commands, interpreting outputs, or troubleshooting. Upstream: https://github.com/grp-bork/gunc. For stage routing use tool-selection / metagenomics-workflow.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/PKU-EMBL/Metagenomics-Skills --skill gunc命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
基于 SOC 职业分类
正在显示 SKILL.md
| name | gunc |
| description | Detect chimerism/contamination in prokaryotic genomes/MAGs. Use this skill when working with gunc: installing, choosing parameters, running commands, interpreting outputs, or troubleshooting. Upstream: https://github.com/grp-bork/gunc. For stage routing use tool-selection / metagenomics-workflow. |
| license | MIT |
| category | analysis-tools |
| tags | ["qc","contamination","chimerism"] |
| upstream | https://github.com/grp-bork/gunc |
| stage | assessment |
Upstream: GUNC
Orakov, A. et al. GUNC: detection of chimerism and contamination in prokaryotic genomes. Genome Biol. 22, 178 (2021). https://doi.org/10.1186/s13059-021-02393-0
See also docs/references.md.
CheckM-style metrics estimate surplus marker copies but under-detect some clade-level chimeras (contigs from distinct lineages merged into one bin). GUNC was built for those underdetected contamination types; the paper conservatively estimated substantial residual contamination in public genomes even after conventional QC.
Use GUNC with CheckM2 before strong claims on novel lineages, GEMs, or BGCs tied to a single MAG. Contamination types to keep distinct in text:
| Type | Rough meaning |
|---|---|
| Redundant | Extra copies from related genome content |
| Non-redundant / chimeric | Foreign-lineage fragments in one “genome” |
Single-coverage binning can hide co-abundant contaminants within one sample — multi-coverage binning + GUNC is a strong pair (Mattock & Watson 2023).
conda install -c bioconda gunc # if packaged; else follow upstream
gunc run -i bins/ --db_file $GUNC_DB -r temp/gunc -t 16
| Item | Notes |
|---|---|
| DB | Download/pin GUNC DB |
| With CheckM2 | Combine filters; discordant calls → manual review |
| Novel lineages | Higher scrutiny; GUNC DB taxonomy limits apply |
checkm2 · mag-qc · drep · basalt · study-design · tool-selection