By name: varcadd-pathogenicity description: Variant Scorer keywords: - variant-interpretation - CADD - pathogenicity - genomics - prediction measurable_outcome: Return pathogenicity scores for a VCF of 1000 variants within 2 minutes, flagging top 1% deleterious hits. license: Non-Commercial metadata: author: Genome Medicine 2025 version: "1.0.0" compatibility: - system: Python 3.9+ allowed-tools: - run_shell_command - read_file
varCADD (Variant Pathogenicity Predictor)
Genome-wide pathogenicity prediction leveraging standing variation data to improve accuracy over traditional CADD scores.
When to Use
- Variant Prioritization: Ranking candidate variants in rare disease cases.
- VUS Interpretation: Assessing variants of uncertain significance.
- Research: Annotating novel variants in population studies.
Core Capabilities
- Score Generation: Calculate C-scores for SNVs and indels.
- Annotation: Add functional context (conservation, protein domains).
- Filtering: Identify likely pathogenic variants based on thresholds.
Workflow
- Input: VCF file.
- Annotate: Run varCADD model.
- Filter: Keep variants with Score > X.
- Output: Annotated VCF or ranked table.
Example Usage
User: "Score these variants from patient X."
Agent Action:
varcadd score --input patient.vcf --output scored.vcf