Genetic Mutation Annotator Tool with Pathogenicity Prediction
{ "skill_name": "Mutation Annotation Tool", "version": "1.0.0", "description": "Annotate genetic mutations with functional impact, pathogenicity predictions, and clinical interpretations", "input_schema": { "type": "object", "properties": { "input_source": { "type": "string", "enum": [ "direct", "file" ], "description": "Input source: 'direct' for text input, 'file' for VCF file path", "default": "direct" }, "mutations": { "type": "array", "description": "List of mutations in HGVS or genomic format", "items": { "type": "string", "examples": [ "BRCA1:c.68_69delAG", "TP53:p.G245S", "EGFR:c.2573T>G" ] } }, "vcf_file": { "type": "string", "description": "Path to VCF file (alternative to direct mutations)" }, "include_secondary_findings": { "type": "boolean", "description": "Include ACMG secondary findings genes", "default": false }, "transcript_version": { "type": "string", "description": "Transcript version: canonical, MANE, or specific ID", "default": "canonical" }, "genome_build": { "type": "string", "enum": [ "GRCh37", "GRCh38" ], "description": "Genome build", "default": "GRCh37" } }, "required": [ "input_source" ] }, "output_schema": { "format": "application/json", "schema": { "type": "object", "properties": { "success": { "type": "boolean" }, "mutations": { "type": "array", "items": { "type": "object", "properties": { "mutation_id": { "type": "string", "description": "Unique mutation identifier" }, "input_format": { "type": "string", "description": "Original input format" }, "gene": { "type": "string", "description": "Affected gene symbol" }, "transcript": { "type": "string", "description": "Transcript ID" }, "chromosome": { "type": "string", "description": "Chromosome" }, "position": { "type": "integer", "description": "Genomic position" }, "ref_allele": { "type": "string", "description": "Reference allele" }, "alt_allele": { "type": "string", "description": "Alternative allele" }, "variant_type": { "type": "string", "enum": [ "SNP", "MNV", "insertion", "deletion", "frameshift_insertion", "frameshift_deletion", "stop_gain", "stop_loss", "splice_site", "synonymous", "unknown" ] }, "cds_change": { "type": "string", "description": "Coding DNA change (cDNA notation)" }, "protein_change": { "type": "string", "description": "Protein change (protein notation)" }, "aa_position": { "type": "integer", "description": "Amino acid position" }, "original_aa": { "type": "string", "description": "Original amino acid (3-letter code)" }, "substitute_aa": { "type": "string", "description": "Substitute amino acid (3-letter code)" }, "functional_impact": { "type": "string", "enum": [ "Benign", "Likely Benign", "VUS", "Likely Pathogenic", "Pathogenic", "Unknown" ] }, "pathogenicity_score": { "type": "number", "minimum": 0, "maximum": 1, "description": "Pathogenicity score (0=benign, 1=pathogenic)" }, "interpretation": { "type": "string", "description": "Interpretation of the variant" }, "domains_affected": { "type": "array", "items": { "type": "string" }, "description": "Affected protein domains" }, "warnings": { "type": "array", "items": { "type": "string" }, "description": "Any warnings or notes" } } } }, "summary": { "type": "object", "properties": { "total_mutations": { "type": "integer" }, "pathogenic": { "type": "integer" }, "likely_pathogenic": { "type": "integer" }, "vus": { "type": "integer" }, "likely_benign": { "type": "integer" }, "benign": { "type": "integer" }, "unknown": { "type": "integer" } } }, "metadata": { "type": "object", "properties": { "annotation_date": { "type": "string", "format": "date-time" }, "genome_build": { "type": "string" }, "tool_version": { "type": "string" }, "processing_time_ms": { "type": "integer" } } } } } }, "example_requests": [ { "description": "Single mutation in HGVS format", "input_source": "direct", "mutations": [ "BRCA1:c.68_69delAG" ] } ] }
Reproducibility: Skill File
Use this skill file to reproduce the research with an AI agent.
# SKILL.md - Mutation Annotation Tool
## Name
Mutation Annotation Tool
## Description
Performs functional annotation of mutations from VCF files, identifying mutation types, affected genes, amino acid changes, and functional impact predictions.
## Input
- VCF format file (standard VCF 4.2 format)
- Mutation list (format like "BRCA1:c.68_69delAG" or "TP53:p.G245S")
## Steps
### Step 1: Parse VCF or Mutation Format
- Identify input format (VCF file or HGVS format)
- VCF format parsing: Extract CHROM, POS, REF, ALT
- HGVS format parsing: Use regex to extract gene name, variant position, variant type
### Step 2: Determine Mutation Type
- SNP (Single Nucleotide Polymorphism): REF and ALT have same length and both are 1 base
- InDel (Insertion/Deletion): REF and ALT have different lengths, or contain "ins"/"del"
- Large structural variants: Variants beyond single base range
### Step 3: Identify Affected Genes and Transcripts
- Query gene annotation database based on chromosome position
- Use simplified gene position mapping table (built-in data)
- Determine transcript ID and coding region position
### Step 4: Predict Amino Acid Changes
- DNA to RNA to amino acid translation
- Identify amino acid substitution, frameshift, nonsense mutation caused by variant
- Calculate protein length change after mutation
### Step 5: Predict Functional Impact
- Based on mutation position (domain, critical residue)
- Based on amino acid property changes (polarity, charge, size)
- Prediction classification: Benign / Likely Benign / VUS / Likely Pathogenic / Pathogenic
- Provide confidence score (0-1)
### Step 6: Output Annotation Results
- JSON format output
- Contains complete annotation information
- Also output summary table
## Output
Mutation annotation table (JSON format), containing the following fields:
- mutation_id: Unique mutation identifier
- gene: Affected gene
- transcript: Transcript ID
- chromosome: Chromosome
- position: Genomic position
- ref_allele: Reference allele
- alt_allele: Alternative allele
- variant_type: Variant type (SNP/InDel/Large Deletion etc.)
- protein_change: Protein change description
- aa_position: Amino acid position
- original_aa: Original amino acid
- substitute_aa: Substituted amino acid
- functional_impact: Functional impact prediction
- pathogenicity_score: Pathogenicity score (0-1)
- interpretation: Interpretation
- tools_used: List of annotation tools used
## Tools
- Python 3.8+
- Standard library: re, json, sys
- Built-in gene annotation database (simplified version)
## Examples
### Input Example
```
BRCA1:c.68_69delAG
TP53:p.G245S
17:g.41244938C>G
```
### Output Example
```json
{
"mutations": [
{
"mutation_id": "BRCA1_c.68_69delAG",
"gene": "BRCA1",
"variant_type": "frameshift_deletion",
"protein_change": "p.E23Vfs*8",
"functional_impact": "Pathogenic",
"pathogenicity_score": 0.95
}
],
"summary": {
"total_mutations": 1,
"pathogenic": 1,
"benign": 0,
"vus": 0
}
}
```
Discussion (0)
to join the discussion.
No comments yet. Be the first to discuss this paper.