skills.rest
ClawBio
ClawBioOFFICIAL
Explore all skills by this author →
WebsiteGitHub

cnv-acmg-classifier

Classify copy-number variants using the ClinGen/ACMG 2019 point framework with five-tier verdicts.

1.1k|257|Updated Feb 25, 2026
Search Tags:#classification#acmg#clinical genomics#CNV#clingen#structural variant#dosage sensitivity
One-click install
npx skills add https://github.com/ClawBio/ClawBio --skill cnv-acmg-classifier
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cnv-acmg-classifier
Source: https://github.com/ClawBio/ClawBio/tree/main/skills/cnv-acmg-classifier
Command: npx skills add https://github.com/ClawBio/ClawBio --skill cnv-acmg-classifier

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Manually scoring copy-number variants against the 19-category ClinGen/ACMG rubric is slow, error-prone, and inconsistent between reviewers. This Skill automates deterministic point scoring for deletions and duplications, producing a transparent five-tier classification with a full evidence trail.

Core Features & Use Cases

  • Automated ACMG scoring: Computes Sections 1-3 (genomic content, dosage-sensitivity overlap, gene count) from CNV coordinates, a dosage-sensitivity map, and a gene model, then sums analyst-supplied Section 4/5 evidence.
  • Flexible inputs: Accepts VCF files with SVTYPE/END or CSV/TSV tables of CNV calls, with optional ClinGen dosage maps and gene models.
  • Reproducible reports: Writes a markdown report, machine-readable JSON, a classifications CSV, and a reproducibility bundle with checksums.
  • Use Case: A clinical genomics analyst receives Sarek/Manta SV calls and needs to know whether a 22q11.2 duplication is pathogenic; the Skill scores it against ClinGen dosage sensitivity and returns Pathogenic with evidence codes 1A, 2A, 5A.

Quick Start

Ask the agent to classify the CNVs in your VCF or CSV file with ACMG, or run the demo to see seven synthetic CNVs classified across all five tiers.

Frequently Asked Questions about cnv-acmg-classifier

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I classify a CNV with ACMG criteria?▼

Provide CNV calls as a VCF with SVTYPE/END or a CSV/TSV with cnv_id, chrom, start, end, and type columns. The skill scores each variant against the ClinGen/ACMG 2019 point framework and returns a five-tier classification from Pathogenic to Benign.

What input formats does CNV classification support?▼

The skill accepts VCF files (including gzipped) with SVTYPE and END in the INFO field, or CSV/TSV tables with cnv_id, chrom, start, end, and type columns. Optional inheritance and case_evidence_points columns feed Sections 4 and 5.

Can this classify SNVs or small indels with ACMG?▼

No. This skill only handles copy-number and structural variants (deletions and duplications spanning genes). SNV/indel ACMG classification belongs to a separate clinical-variant-reporter skill, and variant calling itself belongs upstream.

Does CNV ACMG classification require internet access or extra packages?▼

No. The classifier runs entirely on Python 3.10+ standard library with no third-party packages and no network access. All processing is local, and demo data is bundled for offline testing.

Why is my CNV scored as VUS instead of Pathogenic?▼

Scores are additive across all sections, so a complete 2A deletion inherited from an unaffected parent sums to 1.00 minus 0.30, giving 0.70 which is VUS. Check the evidence codes in the report to see which sections contributed points.

What are the limitations of the bundled dosage map?▼

The bundled dosage map and gene model are small curated demonstration subsets, not authoritative references. For real clinical work, pass the full ClinGen Dosage Sensitivity Map and a Gencode/RefSeq gene model, or true Pathogenic CNVs may be silently downgraded.

Related Skills

Explore More Education & Research Skills→
openaiopenai
5.3k

ngs-dna-germline-variants

Plan and run germline WGS, WES, and panel variant-calling workflows from FASTQ, BAM, or CRAM.

Official
Advanced
ClawBioClawBio
1.1k

profile-report

Generates unified genomic profile reports from uploaded JSON data.

Official
Intermediate
nahisahonahisaho
3

scientific-variant-interpretation

Automate ACMG-aligned clinical interpretation of genetic variants and generate standardized reports.

Community
Advanced
nahisahonahisaho
3

scientific-clingen-curation

Automate ClinGen gene-disease evidence curation across validity, dosage sensitivity, and actionability.

Community
Advanced