[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EDDG9C6H0BZ7FWDZ7VG0QF":3,"skill-same:01M7EDDG9C6H0BZ7FWDZ7VG0QF":66},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":30,"commitSha":31,"skillDirectory":32,"forksCount":14,"repositoryHtmlUrl":33,"repositoryHomepageUrl":33,"ownerHtmlUrl":34,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":35,"install":36,"resources":49},"01M7EDDG9C6H0BZ7FWDZ7VG0QF","bio-genome-annotation-annotation-qc",null,"Assesses the quality and completeness of a genome annotation with BUSCO (conserved single-copy ortholog recovery), OMArk (proteome completeness, consistency, and contamination), CheckM2 (prokaryotic completeness\u002Fcontamination), and a gene-set sanity panel (gene count, mono-exonic fraction, protein-length distribution, mRNA:gene ratio, coding density). Covers the assembly-BUSCO-vs-proteome-BUSCO diagnostic, what BUSCO-Duplicated really means, why gene count is a vanity metric, and the QC of transferred annotations. Use when judging whether an annotation is good enough to publish or submit, diagnosing a suspect annotation, or comparing annotation completeness across pipelines.","Mr-Nobody-Anonymous\u002FAll-skills","Mr-Nobody-Anonymous","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F184253119?v=4","01M7AEXFJHKTYQW5R0ZCQQTWFV","anonymous","01M7C00M3GQR4QKP9GZA42GFAK",0,"MIT","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fbiology\u002Fbio-genome-annotation-annotation-qc\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: bio-genome-annotation-annotation-qc\ndescription: Assesses the quality and completeness of a genome annotation with BUSCO (conserved single-copy ortholog recovery), OMArk (proteome completeness, consistency, and contamination), CheckM2 (prokaryotic completeness\u002Fcontamination), and a gene-set sanity panel (gene count, mono-exonic fraction, protein-length distribution, mRNA:gene ratio, coding density). Covers the assembly-BUSCO-vs-proteome-BUSCO diagnostic, what BUSCO-Duplicated really means, why gene count is a vanity metric, and the QC of transferred annotations. Use when judging whether an annotation is good enough to publish or submit, diagnosing a suspect annotation, or comparing annotation completeness across pipelines.\ntool_type: cli\nprimary_tool: BUSCO\nsource: \"https:\u002F\u002Fgithub.com\u002FGPTomics\u002FbioSkills\"\nsource_repository: \"GPTomics\u002FbioSkills\"\nsource_path: \"genome-annotation\u002Fannotation-qc\u002FSKILL.md\"\nlicense: \"MIT\"\nimported_at: \"2026-09-20\"\n---\n\n## Version Compatibility\n\nReference examples tested with: BUSCO 5.5+, OMArk 0.3+, CheckM2 1.0+, compleasm 0.2.6+, gffutils 0.12+, matplotlib 3.8+.\n\nBefore using code patterns, verify installed versions match. If versions differ:\n- CLI: `\u003Ctool> --version` then `\u003Ctool> --help` to confirm flags\n- Python: `pip show \u003Cpackage>` then `help(module.function)` to check signatures\n\nBUSCO results depend on the **lineage dataset** (e.g. `vertebrata_odb10` vs the shallow `eukaryota_odb10`) and the OrthoDB version - record them; a 99% on a shallow ~255-gene set and a 99% on a deep ~5,500-gene clade set are very different claims. If code throws an error, introspect the installed tool and adapt rather than retrying.\n\n# Annotation QC\n\n**\"Is my genome annotation any good?\"** -> Measure conserved-gene completeness, proteome consistency\u002Fcontamination, and gene-set sanity, and decide whether the limiting factor is the assembly or the predictor.\n- CLI: `busco -i proteins.faa -m proteins -l \u003Clineage>_odb10` (proteome) and `busco -i genome.fa -m genome -l \u003Clineage>_odb10` (assembly), `omark`, `checkm2 predict` (prokaryotes)\n\n## The Single Most Important Modern Insight -- BUSCO Measures the Easy 10%; the Diagnostic Is Genome-vs-Proteome\n\nBUSCO completeness measures recovery of the ~1,000-5,500 most conserved single-copy orthologs - the most conserved, highly expressed, intron-stable genes any half-broken pipeline will find. A 98%-complete BUSCO confirms the housekeeping core is intact; it says **nothing** about whether the other ~20,000 models are chimeric, fragmented, frame-shifted, fused, or hallucinated from TEs. BUSCO is a smoke detector in one room of a burning house. Three load-bearing moves:\n\n1. **Run assembly-BUSCO vs proteome-BUSCO on the same assembly - the diagnostic almost nobody runs.** `-m genome` does its own gene-finding; `-m proteins` scores the delivered proteome. If assembly is 98% and proteome is 85%, the genes are physically present and the *predictor* missed them - a training\u002Fevidence\u002Fmasking problem, fixable without touching the assembly. If both are low, the genes aren't in the assembly - stop annotating and fix the assembly. This single fork redirects more wasted effort than any other check. (Always run BUSCO on the *delivered proteome*, `-m proteins`, not genome mode reported as if it described the annotation.)\n2. **Gene count is a vanity metric.** The *same* assembly annotated by two pipelines routinely differs 20-40% in gene count, and a higher count is as likely to mean spurious TE ORFs and split models as more real genes. The diagnostic signal is in the **mono-exonic fraction, protein-length distribution, mean exons\u002Fgene, and mRNA:gene ratio** - not the headline count.\n3. **Complement BUSCO with OMArk.** BUSCO's single-copy-ortholog lens is blind to over-prediction, chimeras, and contamination; OMArk assesses the *whole* proteome for completeness AND consistency (are genes consistent with their homologs?) and detects contaminant species. Use both.\n\n## Tool Taxonomy\n\n| Tool | Citation | Measures | Domain |\n|------|----------|----------|--------|\n| BUSCO | Manni 2021 *Mol Biol Evol* | conserved single-copy ortholog recovery (C\u002FD\u002FF\u002FM) | euk + prok + viral; proteome\u002Fgenome\u002Ftranscriptome modes |\n| OMArk | Nevers 2025 *Nat Biotechnol* | proteome completeness + consistency + contamination | eukaryotic proteomes; catches over-prediction BUSCO misses |\n| compleasm | Huang 2023 | faster miniprot-based BUSCO-compatible completeness | euk; genome mode |\n| CheckM2 | Chklovski 2023 *Nat Methods* | completeness + contamination (ML, lineage-agnostic) | bacteria\u002Farchaea, isolates + MAGs |\n| Gene-set sanity panel | (gffutils) | gene count, mono-exonic %, protein length, mRNA:gene, coding density | any; the panel BUSCO can't provide |\n| LAI | Ou 2018 *NAR* | LTR-RT assembly resolution (repeat-space contiguity) | LTR-rich genomes; assembly-side QC |\n\n## Decision Tree by Scenario\n\n| Scenario | QC to run | Why |\n|----------|-----------|-----|\n| Prokaryotic genome\u002FMAG | CheckM2 (completeness\u002Fcontamination) + coding density + tRNA\u002FrRNA counts | CheckM2 is the field-standard completeness call; BUSCO is conservative on prokaryotes |\n| Eukaryotic annotation | BUSCO `-m proteins` AND `-m genome` + OMArk + gene-set sanity panel | the genome-vs-proteome fork + over-prediction\u002Fcontamination |\n| Suspect high gene count | mono-exonic fraction + protein-length + BUSCO-Duplicated | distinguish haplotigs \u002F unmasked TEs \u002F split models |\n| High BUSCO-Duplicated | check synteny + clade ploidy | uncollapsed haplotigs (purge) vs real WGD (keep) |\n| Transferred annotation | BUSCO on the lifted set vs reference + ORF integrity | quantify silently lost conserved genes |\n| Low proteome-BUSCO, high genome-BUSCO | fix evidence\u002Fmasking\u002Ftraining | the genes are present; the predictor missed them |\n| Both BUSCO low | -> genome-assembly\u002Fassembly-qc (purge_dups, more data) | the limit is the assembly |\n\n## BUSCO and How to Read It\n\n```bash\nbusco -i proteins.faa -m proteins -l vertebrata_odb10 -o busco_prot -c 16   # the delivered proteome\nbusco -i genome.fa    -m genome   -l vertebrata_odb10 -o busco_genome -c 16 # the assembly\n```\n\nReported as `C:[S,D],F,M`: **Complete** (full-length match), **Duplicated** (subset of Complete, found ≥2x), **Fragmented** (partial), **Missing**. Use the **deepest applicable clade dataset**, not the shallow `eukaryota_odb10` (a 99% on a 255-gene set is trivially easy and not comparable to a 99% on a 5,500-gene clade set). **High Duplicated is the most-misread signal** - three causes, opposite responses: uncollapsed haplotigs (genome-wide, heterozygosity-scaled, non-syntenic -> purge_dups *before* annotating), real recent WGD (syntenic, ploidy-consistent, Ks duplication peak -> keep), or split models from a fragmented assembly. A clean haploid annotation runs ~1-3% Duplicated; >5-8% with no known WGD screams purge_dups first. compleasm is a faster miniprot-based alternative that often reports higher completeness than BUSCO's metaeuk path.\n\n## OMArk (Proteome Consistency and Contamination)\n\n```bash\nomamer search --db LUCA.h5 --query proteins.faa --out proteins.omamer\nomark -f proteins.omamer -d LUCA.h5 -o omark_out\n```\n\nOMArk catches what BUSCO's single-copy lens misses: it classifies the *whole* proteome as consistent \u002F inconsistent (a gene whose structure conflicts with its homologs - a chimera or fragment) \u002F unknown, and detects **contaminant species** mixed into the proteome. A high \"inconsistent\" fraction flags over-prediction or fused\u002Fsplit models even when BUSCO is green.\n\n## CheckM2 (Prokaryotic Completeness\u002FContamination)\n\n```bash\ncheckm2 predict --input genome.fna --output-directory checkm2_out --threads 16\n```\n\nFor bacteria\u002Farchaea, CheckM2's lineage-agnostic ML model is the standard completeness\u002Fcontamination call (handles reduced\u002Fnovel lineages where marker sets fail). Gate annotation on it: **contamination >5%** mixes two organisms' genes into a chimeric set; **completeness \u003C90% with contamination >5-10%** makes gene count, coding density, and hypothetical fraction uninterpretable - fix the assembly\u002Fbinning first.\n\n## Gene-Set Sanity Panel with Python\n\n**Goal:** Compute the triage panel that reveals annotation health where gene count and BUSCO cannot.\n\n**Approach:** Load the GFF3 into gffutils; compute the mRNA:gene ratio (1.00 = isoform\u002FUTR-naive), the mono-exonic fraction, and the protein-length distribution; flag clade-anomalous values.\n\n```python\nimport gffutils\n\nMONOEXONIC_FLAG = 0.30   # >30% single-exon in a vertebrate suggests unmasked TEs\u002Fpseudogenes\u002Ffragments (calibrate per clade; fungi run higher)\n\ndef sanity_panel(gff_file):\n    db = gffutils.create_db(gff_file, ':memory:', merge_strategy='merge')\n    genes = list(db.features_of_type('gene'))\n    mrnas = list(db.features_of_type(['mRNA', 'transcript']))\n    exon_counts = [len(list(db.children(tx, featuretype='exon'))) for tx in mrnas]\n    mono_frac = sum(1 for e in exon_counts if e == 1) \u002F len(exon_counts) if exon_counts else 0\n    mrna_per_gene = len(mrnas) \u002F len(genes) if genes else 0\n    if mrna_per_gene \u003C= 1.001:\n        print('WARNING: mRNA:gene == 1.00 -- isoform\u002FUTR-naive; AS\u002F3-prime-tag scRNA-seq analyses untrustworthy')\n    if mono_frac > MONOEXONIC_FLAG:\n        print(f'WARNING: mono-exonic fraction {mono_frac:.1%} high -- check masking\u002Fcontamination')\n    return {'genes': len(genes), 'mrna_per_gene': mrna_per_gene, 'mono_exonic_fraction': mono_frac}\n```\n\nRead gene count against the *nearest well-annotated relative* and the species' ploidy, never in isolation (1.5-2x with no WGD = haplotigs\u002Fover-prediction; ~0.5x = over-masking\u002Funder-training). A healthy protein-length distribution is unimodal near the clade-typical ~300-450 aa; a sub-100-aa spike = spurious\u002Ffragmented calls; a fat left tail = partials from a fragmented assembly.\n\n## Per-Method Failure Modes\n\n### Reporting genome-mode BUSCO as the annotation's score\n**Trigger:** running BUSCO `-m genome` and citing it as annotation quality. **Mechanism:** genome mode does its own gene-finding, often better than the author's pipeline on the conserved core. **Symptom:** reported BUSCO higher than the real proteome BUSCO. **Fix:** run `-m proteins` on the delivered proteome; report both and compare.\n\n### High Duplicated read as success\n**Trigger:** treating high BUSCO-D as good. **Mechanism:** uncollapsed haplotigs vs real WGD vs split models. **Symptom:** inflated gene count. **Fix:** plot duplicated pairs against synteny + clade ploidy; if no WGD and D>5-8%, purge_dups the assembly first.\n\n### Shallow lineage dataset\n**Trigger:** `eukaryota_odb10` (~255 genes) instead of the deep clade set. **Mechanism:** a small, conserved set is trivially easy to hit 99%. **Symptom:** misleadingly high completeness. **Fix:** use the deepest applicable clade dataset; record it.\n\n### BUSCO alone, no OMArk\n**Trigger:** judging a proteome by BUSCO completeness only. **Mechanism:** the single-copy-ortholog lens misses over-prediction, chimeras, and contamination. **Symptom:** green BUSCO on a proteome full of fused\u002Ffragmented or contaminant models. **Fix:** add OMArk for consistency + contamination.\n\n### Annotating before the QC gate (prokaryote)\n**Trigger:** Bakta\u002FProkka before CheckM2. **Mechanism:** contamination mixes organisms; low completeness truncates. **Symptom:** chimeric\u002Finflated gene set, uninterpretable coding density. **Fix:** CheckM2 first; contamination >5% -> decontaminate.\n\n## Quantitative Thresholds\n\n| Threshold | Source | Rationale |\n|-----------|--------|-----------|\n| Assembly-BUSCO vs proteome-BUSCO gap | the diagnostic fork | large gap = predictor missed present genes; both low = fix assembly |\n| BUSCO-Duplicated ~1-3% (clean haploid); >5-8% no WGD | assembly norm | purge_dups before annotating |\n| Use deepest applicable clade `_odb10` | BUSCO guidance | shallow sets trivially hit 99% |\n| Mono-exonic ~10-20% (vertebrate); >25-30% flag | clade norm | unmasked TEs\u002Fpseudogenes\u002Ffragments; fungi legitimately higher |\n| Protein length unimodal ~300-450 aa | eukaryote norm | sub-100-aa spike = spurious; fat left tail = partials |\n| mRNA:gene ratio == 1.00 | annotation structure | isoform\u002FUTR-naive |\n| CheckM2 contamination ≤5%, completeness ≥90% | MIMAG-aligned | above\u002Fbelow -> prokaryotic QC numbers uninterpretable |\n| Prokaryotic coding density ~88-90% | bacterial norm | \u003C85% wrong table\u002Ffragmentation; >93% over-calling |\n\n## Common Errors\n\n| Error \u002F symptom | Cause | Solution |\n|-----------------|-------|----------|\n| BUSCO high, annotation still bad | BUSCO measures only the conserved core | add OMArk + sanity panel; inspect mono-exonic models |\n| Proteome-BUSCO \u003C genome-BUSCO | predictor missed present genes | fix evidence\u002Fmasking\u002Ftraining, not the assembly |\n| High Duplicated | haplotigs vs WGD vs split models | synteny + ploidy check; purge_dups if no WGD |\n| 99% completeness looks too good | shallow lineage dataset | rerun with the deep clade set |\n| Cross-pipeline gene counts disagree | gene count is not a quality metric | compare sanity panels, not counts |\n| CheckM2 high contamination | mixed organisms \u002F poor binning | decontaminate before annotation |\n\n## References\n\n- Manni M, et al. 2021. BUSCO update: novel and streamlined workflows along with broader and deeper phylogenetic coverage for scoring of eukaryotic, prokaryotic, and viral genomes. *Mol Biol Evol* 38:4647-4654.\n- Simão FA, et al. 2015. BUSCO: assessing genome assembly and annotation completeness with single-copy orthologs. *Bioinformatics* 31:3210-3212.\n- Nevers Y, et al. 2025. Quality assessment of gene repertoire annotations with OMArk. *Nat Biotechnol* 43:124-133.\n- Chklovski A, et al. 2023. CheckM2: a rapid, scalable and accurate tool for assessing microbial genome quality using machine learning. *Nat Methods* 20:1203-1212.\n- Huang N, Li H. 2023. compleasm: a faster and more accurate reimplementation of BUSCO. *Bioinformatics* 39:btad595.\n- Guan D, et al. 2020. Identifying and removing haplotypic duplication in primary genome assemblies (purge_dups). *Bioinformatics* 36:2896-2898.\n- Ou S, Chen J, Jiang N. 2018. Assessing genome assembly quality using the LTR Assembly Index (LAI). *Nucleic Acids Res* 46:e126.\n\n## Related Skills\n\n- eukaryotic-gene-prediction - The annotation whose proteome\u002Fgene-set this QC evaluates\n- prokaryotic-annotation - CheckM2 gate and coding-density sanity for prokaryotes\n- annotation-transfer - BUSCO on a lifted set quantifies silently lost conserved genes\n- repeat-annotation - LAI and over-masking diagnosis in the assembly-to-annotation handoff\n- genome-assembly\u002Fassembly-qc - Assembly-side completeness; purge haplotigs before annotating\n","## Version Compatibility\n\nReference examples tested with: BUSCO 5.5+, OMArk 0.3+, CheckM2 1.0+, compleasm 0.2.6+, gffutils 0.12+, matplotlib 3.8+.\n\nBefore using code patterns, verify installed versions match. If versions differ:\n- CLI: `\u003Ctool> --version` then `\u003Ctool> --help` to confirm flags\n- Python: `pip show \u003Cpackage>` then `help(module.function)` to check signatures\n\nBUSCO results depend on the **lineage dataset** (e.g. `vertebrata_odb10` vs the shallow `eukaryota_odb10`) and the OrthoDB version - record them; a 99% on a shallow ~255-gene set and a 99% on a deep ~5,500-gene clade set are very different claims. If code throws an error, introspect the installed tool and adapt rather than retrying.\n\n# Annotation QC\n\n**\"Is my genome annotation any good?\"** -> Measure conserved-gene completeness, proteome consistency\u002Fcontamination, and gene-set sanity, and decide whether the limiting factor is the assembly or the predictor.\n- CLI: `busco -i proteins.faa -m proteins -l \u003Clineage>_odb10` (proteome) and `busco -i genome.fa -m genome -l \u003Clineage>_odb10` (assembly), `omark`, `checkm2 predict` (prokaryotes)\n\n## The Single Most Important Modern Insight -- BUSCO Measures the Easy 10%; the Diagnostic Is Genome-vs-Proteome\n\nBUSCO completeness measures recovery of the ~1,000-5,500 most conserved single-copy orthologs - the most conserved, highly expressed, intron-stable genes any half-broken pipeline will find. A 98%-complete BUSCO confirms the housekeeping core is intact; it says **nothing** about whether the other ~20,000 models are chimeric, fragmented, frame-shifted, fused, or hallucinated from TEs. BUSCO is a smoke detector in one room of a burning house. Three load-bearing moves:\n\n1. **Run assembly-BUSCO vs proteome-BUSCO on the same assembly - the diagnostic almost nobody runs.** `-m genome` does its own gene-finding; `-m proteins` scores the delivered proteome. If assembly is 98% and proteome is 85%, the genes are physically present and the *predictor* missed them - a training\u002Fevidence\u002Fmasking problem, fixable without touching the assembly. If both are low, the genes aren't in the assembly - stop annotating and fix the assembly. This single fork redirects more wasted effort than any other check. (Always run BUSCO on the *delivered proteome*, `-m proteins`, not genome mode reported as if it described the annotation.)\n2. **Gene count is a vanity metric.** The *same* assembly annotated by two pipelines routinely differs 20-40% in gene count, and a higher count is as likely to mean spurious TE ORFs and split models as more real genes. The diagnostic signal is in the **mono-exonic fraction, protein-length distribution, mean exons\u002Fgene, and mRNA:gene ratio** - not the headline count.\n3. **Complement BUSCO with OMArk.** BUSCO's single-copy-ortholog lens is blind to over-prediction, chimeras, and contamination; OMArk assesses the *whole* proteome for completeness AND consistency (are genes consistent with their homologs?) and detects contaminant species. Use both.\n\n## Tool Taxonomy\n\n| Tool | Citation | Measures | Domain |\n|------|----------|----------|--------|\n| BUSCO | Manni 2021 *Mol Biol Evol* | conserved single-copy ortholog recovery (C\u002FD\u002FF\u002FM) | euk + prok + viral; proteome\u002Fgenome\u002Ftranscriptome modes |\n| OMArk | Nevers 2025 *Nat Biotechnol* | proteome completeness + consistency + contamination | eukaryotic proteomes; catches over-prediction BUSCO misses |\n| compleasm | Huang 2023 | faster miniprot-based BUSCO-compatible completeness | euk; genome mode |\n| CheckM2 | Chklovski 2023 *Nat Methods* | completeness + contamination (ML, lineage-agnostic) | bacteria\u002Farchaea, isolates + MAGs |\n| Gene-set sanity panel | (gffutils) | gene count, mono-exonic %, protein length, mRNA:gene, coding density | any; the panel BUSCO can't provide |\n| LAI | Ou 2018 *NAR* | LTR-RT assembly resolution (repeat-space contiguity) | LTR-rich genomes; assembly-side QC |\n\n## Decision Tree by Scenario\n\n| Scenario | QC to run | Why |\n|----------|-----------|-----|\n| Prokaryotic genome\u002FMAG | CheckM2 (completeness\u002Fcontamination) + coding density + tRNA\u002FrRNA counts | CheckM2 is the field-standard completeness call; BUSCO is conservative on prokaryotes |\n| Eukaryotic annotation | BUSCO `-m proteins` AND `-m genome` + OMArk + gene-set sanity panel | the genome-vs-proteome fork + over-prediction\u002Fcontamination |\n| Suspect high gene count | mono-exonic fraction + protein-length + BUSCO-Duplicated | distinguish haplotigs \u002F unmasked TEs \u002F split models |\n| High BUSCO-Duplicated | check synteny + clade ploidy | uncollapsed haplotigs (purge) vs real WGD (keep) |\n| Transferred annotation | BUSCO on the lifted set vs reference + ORF integrity | quantify silently lost conserved genes |\n| Low proteome-BUSCO, high genome-BUSCO | fix evidence\u002Fmasking\u002Ftraining | the genes are present; the predictor missed them |\n| Both BUSCO low | -> genome-assembly\u002Fassembly-qc (purge_dups, more data) | the limit is the assembly |\n\n## BUSCO and How to Read It\n\n```bash\nbusco -i proteins.faa -m proteins -l vertebrata_odb10 -o busco_prot -c 16   # the delivered proteome\nbusco -i genome.fa    -m genome   -l vertebrata_odb10 -o busco_genome -c 16 # the assembly\n```\n\nReported as `C:[S,D],F,M`: **Complete** (full-length match), **Duplicated** (subset of Complete, found ≥2x), **Fragmented** (partial), **Missing**. Use the **deepest applicable clade dataset**, not the shallow `eukaryota_odb10` (a 99% on a 255-gene set is trivially easy and not comparable to a 99% on a 5,500-gene clade set). **High Duplicated is the most-misread signal** - three causes, opposite responses: uncollapsed haplotigs (genome-wide, heterozygosity-scaled, non-syntenic -> purge_dups *before* annotating), real recent WGD (syntenic, ploidy-consistent, Ks duplication peak -> keep), or split models from a fragmented assembly. A clean haploid annotation runs ~1-3% Duplicated; >5-8% with no known WGD screams purge_dups first. compleasm is a faster miniprot-based alternative that often reports higher completeness than BUSCO's metaeuk path.\n\n## OMArk (Proteome Consistency and Contamination)\n\n```bash\nomamer search --db LUCA.h5 --query proteins.faa --out proteins.omamer\nomark -f proteins.omamer -d LUCA.h5 -o omark_out\n```\n\nOMArk catches what BUSCO's single-copy lens misses: it classifies the *whole* proteome as consistent \u002F inconsistent (a gene whose structure conflicts with its homologs - a chimera or fragment) \u002F unknown, and detects **contaminant species** mixed into the proteome. A high \"inconsistent\" fraction flags over-prediction or fused\u002Fsplit models even when BUSCO is green.\n\n## CheckM2 (Prokaryotic Completeness\u002FContamination)\n\n```bash\ncheckm2 predict --input genome.fna --output-directory checkm2_out --threads 16\n```\n\nFor bacteria\u002Farchaea, CheckM2's lineage-agnostic ML model is the standard completeness\u002Fcontamination call (handles reduced\u002Fnovel lineages where marker sets fail). Gate annotation on it: **contamination >5%** mixes two organisms' genes into a chimeric set; **completeness \u003C90% with contamination >5-10%** makes gene count, coding density, and hypothetical fraction uninterpretable - fix the assembly\u002Fbinning first.\n\n## Gene-Set Sanity Panel with Python\n\n**Goal:** Compute the triage panel that reveals annotation health where gene count and BUSCO cannot.\n\n**Approach:** Load the GFF3 into gffutils; compute the mRNA:gene ratio (1.00 = isoform\u002FUTR-naive), the mono-exonic fraction, and the protein-length distribution; flag clade-anomalous values.\n\n```python\nimport gffutils\n\nMONOEXONIC_FLAG = 0.30   # >30% single-exon in a vertebrate suggests unmasked TEs\u002Fpseudogenes\u002Ffragments (calibrate per clade; fungi run higher)\n\ndef sanity_panel(gff_file):\n    db = gffutils.create_db(gff_file, ':memory:', merge_strategy='merge')\n    genes = list(db.features_of_type('gene'))\n    mrnas = list(db.features_of_type(['mRNA', 'transcript']))\n    exon_counts = [len(list(db.children(tx, featuretype='exon'))) for tx in mrnas]\n    mono_frac = sum(1 for e in exon_counts if e == 1) \u002F len(exon_counts) if exon_counts else 0\n    mrna_per_gene = len(mrnas) \u002F len(genes) if genes else 0\n    if mrna_per_gene \u003C= 1.001:\n        print('WARNING: mRNA:gene == 1.00 -- isoform\u002FUTR-naive; AS\u002F3-prime-tag scRNA-seq analyses untrustworthy')\n    if mono_frac > MONOEXONIC_FLAG:\n        print(f'WARNING: mono-exonic fraction {mono_frac:.1%} high -- check masking\u002Fcontamination')\n    return {'genes': len(genes), 'mrna_per_gene': mrna_per_gene, 'mono_exonic_fraction': mono_frac}\n```\n\nRead gene count against the *nearest well-annotated relative* and the species' ploidy, never in isolation (1.5-2x with no WGD = haplotigs\u002Fover-prediction; ~0.5x = over-masking\u002Funder-training). A healthy protein-length distribution is unimodal near the clade-typical ~300-450 aa; a sub-100-aa spike = spurious\u002Ffragmented calls; a fat left tail = partials from a fragmented assembly.\n\n## Per-Method Failure Modes\n\n### Reporting genome-mode BUSCO as the annotation's score\n**Trigger:** running BUSCO `-m genome` and citing it as annotation quality. **Mechanism:** genome mode does its own gene-finding, often better than the author's pipeline on the conserved core. **Symptom:** reported BUSCO higher than the real proteome BUSCO. **Fix:** run `-m proteins` on the delivered proteome; report both and compare.\n\n### High Duplicated read as success\n**Trigger:** treating high BUSCO-D as good. **Mechanism:** uncollapsed haplotigs vs real WGD vs split models. **Symptom:** inflated gene count. **Fix:** plot duplicated pairs against synteny + clade ploidy; if no WGD and D>5-8%, purge_dups the assembly first.\n\n### Shallow lineage dataset\n**Trigger:** `eukaryota_odb10` (~255 genes) instead of the deep clade set. **Mechanism:** a small, conserved set is trivially easy to hit 99%. **Symptom:** misleadingly high completeness. **Fix:** use the deepest applicable clade dataset; record it.\n\n### BUSCO alone, no OMArk\n**Trigger:** judging a proteome by BUSCO completeness only. **Mechanism:** the single-copy-ortholog lens misses over-prediction, chimeras, and contamination. **Symptom:** green BUSCO on a proteome full of fused\u002Ffragmented or contaminant models. **Fix:** add OMArk for consistency + contamination.\n\n### Annotating before the QC gate (prokaryote)\n**Trigger:** Bakta\u002FProkka before CheckM2. **Mechanism:** contamination mixes organisms; low completeness truncates. **Symptom:** chimeric\u002Finflated gene set, uninterpretable coding density. **Fix:** CheckM2 first; contamination >5% -> decontaminate.\n\n## Quantitative Thresholds\n\n| Threshold | Source | Rationale |\n|-----------|--------|-----------|\n| Assembly-BUSCO vs proteome-BUSCO gap | the diagnostic fork | large gap = predictor missed present genes; both low = fix assembly |\n| BUSCO-Duplicated ~1-3% (clean haploid); >5-8% no WGD | assembly norm | purge_dups before annotating |\n| Use deepest applicable clade `_odb10` | BUSCO guidance | shallow sets trivially hit 99% |\n| Mono-exonic ~10-20% (vertebrate); >25-30% flag | clade norm | unmasked TEs\u002Fpseudogenes\u002Ffragments; fungi legitimately higher |\n| Protein length unimodal ~300-450 aa | eukaryote norm | sub-100-aa spike = spurious; fat left tail = partials |\n| mRNA:gene ratio == 1.00 | annotation structure | isoform\u002FUTR-naive |\n| CheckM2 contamination ≤5%, completeness ≥90% | MIMAG-aligned | above\u002Fbelow -> prokaryotic QC numbers uninterpretable |\n| Prokaryotic coding density ~88-90% | bacterial norm | \u003C85% wrong table\u002Ffragmentation; >93% over-calling |\n\n## Common Errors\n\n| Error \u002F symptom | Cause | Solution |\n|-----------------|-------|----------|\n| BUSCO high, annotation still bad | BUSCO measures only the conserved core | add OMArk + sanity panel; inspect mono-exonic models |\n| Proteome-BUSCO \u003C genome-BUSCO | predictor missed present genes | fix evidence\u002Fmasking\u002Ftraining, not the assembly |\n| High Duplicated | haplotigs vs WGD vs split models | synteny + ploidy check; purge_dups if no WGD |\n| 99% completeness looks too good | shallow lineage dataset | rerun with the deep clade set |\n| Cross-pipeline gene counts disagree | gene count is not a quality metric | compare sanity panels, not counts |\n| CheckM2 high contamination | mixed organisms \u002F poor binning | decontaminate before annotation |\n\n## References\n\n- Manni M, et al. 2021. BUSCO update: novel and streamlined workflows along with broader and deeper phylogenetic coverage for scoring of eukaryotic, prokaryotic, and viral genomes. *Mol Biol Evol* 38:4647-4654.\n- Simão FA, et al. 2015. BUSCO: assessing genome assembly and annotation completeness with single-copy orthologs. *Bioinformatics* 31:3210-3212.\n- Nevers Y, et al. 2025. Quality assessment of gene repertoire annotations with OMArk. *Nat Biotechnol* 43:124-133.\n- Chklovski A, et al. 2023. CheckM2: a rapid, scalable and accurate tool for assessing microbial genome quality using machine learning. *Nat Methods* 20:1203-1212.\n- Huang N, Li H. 2023. compleasm: a faster and more accurate reimplementation of BUSCO. *Bioinformatics* 39:btad595.\n- Guan D, et al. 2020. Identifying and removing haplotypic duplication in primary genome assemblies (purge_dups). *Bioinformatics* 36:2896-2898.\n- Ou S, Chen J, Jiang N. 2018. Assessing genome assembly quality using the LTR Assembly Index (LAI). *Nucleic Acids Res* 46:e126.\n\n## Related Skills\n\n- eukaryotic-gene-prediction - The annotation whose proteome\u002Fgene-set this QC evaluates\n- prokaryotic-annotation - CheckM2 gate and coding-density sanity for prokaryotes\n- annotation-transfer - BUSCO on a lifted set quantifies silently lost conserved genes\n- repeat-annotation - LAI and over-masking diagnosis in the assembly-to-annotation handoff\n- genome-assembly\u002Fassembly-qc - Assembly-side completeness; purge haplotigs before annotating\n",{"name":5,"source":24,"license":15,"tool_type":25,"description":7,"imported_at":26,"source_path":27,"primary_tool":28,"source_repository":29},"https:\u002F\u002Fgithub.com\u002FGPTomics\u002FbioSkills","cli","2026-09-20","genome-annotation\u002Fannotation-qc\u002FSKILL.md","BUSCO","GPTomics\u002FbioSkills","main","2d1750717747097b602acf30bd09a7cbd4b4c6ea","awesome_skills\u002Fbiology\u002Fbio-genome-annotation-annotation-qc","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous","2026-11-09 11:45:36",{"source":8,"skillName":5,"commands":37,"prompt":47,"zipAvailable":48},[38,41,44],{"tool":39,"command":40},"npx","npx skills add Mr-Nobody-Anonymous\u002FAll-skills --skill bio-genome-annotation-annotation-qc",{"tool":42,"command":43},"bunx","bunx skills add Mr-Nobody-Anonymous\u002FAll-skills --skill bio-genome-annotation-annotation-qc",{"tool":45,"command":46},"pnpm","pnpm dlx skills add Mr-Nobody-Anonymous\u002FAll-skills --skill bio-genome-annotation-annotation-qc","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fbiology\u002Fbio-genome-annotation-annotation-qc\u002FSKILL.md",true,[50,56,60],{"path":51,"resourceType":52,"resourceTypeDesc":53,"sizeBytes":54,"sourceHtmlUrl":55},"examples\u002Fbusco_genome_vs_proteome.sh","SCRIPT","脚本",1720,"https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fbiology\u002Fbio-genome-annotation-annotation-qc\u002Fexamples\u002Fbusco_genome_vs_proteome.sh",{"path":57,"resourceType":52,"resourceTypeDesc":53,"sizeBytes":58,"sourceHtmlUrl":59},"examples\u002Fgene_set_sanity.py",3153,"https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fbiology\u002Fbio-genome-annotation-annotation-qc\u002Fexamples\u002Fgene_set_sanity.py",{"path":61,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":65},"usage-guide.md","REFERENCE","参考文档",4280,"https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fbiology\u002Fbio-genome-annotation-annotation-qc\u002Fusage-guide.md",[67,75,82,89,96,103],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":74},"01M7FFFZPC2025JQQHPHDX154J","dei-assessment","Assess diversity, equity, and inclusion with metrics. TRIGGERS - Use when user needs help with dei-assessment related tasks.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fdei-assessment\u002FSKILL.md",[],[],"2026-11-16 10:05:34",{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":74},"01M7FFFWN8H4WA7A6Y398QKB6J","dei-action-plan","Create DEI action plans with goals and initiatives. TRIGGERS - Use when user needs help with dei-action-plan related tasks.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fdei-action-plan\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":74},"01M7FFFSHTM0QZRVZ2NRJTJ0PY","culture-fit-assessment","Design culture add assessment processes. TRIGGERS - Use when user needs help with culture-fit-assessment related tasks.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fculture-fit-assessment\u002FSKILL.md",[],[],{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":74},"01M7FFFMA12RKEQA38KHJSXXPS","cowork-to-code-bridge","Use an already-installed, independently verified cowork-to-code bridge to run narrowly approved actions on the user's own macOS, Linux, or WSL2 machine through a local file queue.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fcowork-to-code-bridge\u002FSKILL.md",[],[],{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"publishedAt":74},"01M7FFFH1F6M8YRGW5V487GH1K","counter-offer-response","Draft counter-offer responses for retention. TRIGGERS - Use when user needs help with counter-offer-response related tasks.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fcounter-offer-response\u002FSKILL.md",[],[],{"skillNo":104,"name":105,"nameZh":6,"description":106,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":107,"categories":108,"keywords":109,"publishedAt":74},"01M7FFFDGEHV8Y7J9N40HZXW52","continuous-feedback-system","Design continuous feedback systems. TRIGGERS - Use when user needs help with continuous-feedback-system related tasks.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fcontinuous-feedback-system\u002FSKILL.md",[],[]]