assess_pathogenicity
Evaluate genetic variant pathogenicity with clinical classification, scoring, and evidence breakdown for diagnostic sequencing and variant interpretation.
Instructions
Comprehensive pathogenicity assessment of a genetic variant.
Evaluates variant across all regulatory modalities and provides clinical classification.
Returns:
Pathogenicity score (0-1 scale)
Clinical classification (pathogenic/likely_pathogenic/uncertain/likely_benign/benign)
Evidence breakdown (expression, splicing, TF binding impacts)
Perfect for: clinical variant interpretation, pathogenicity prediction, diagnostic sequencing.
Example: "Assess pathogenicity of chr19:44908684T>C"
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| chromosome | Yes | Chromosome (chr1-chr22, chrX, chrY) | |
| position | Yes | Genomic position (1-based) | |
| ref | Yes | Reference allele | |
| alt | Yes | Alternate allele | |
| tissue_type | No | Optional: disease-relevant tissue (default: brain) |
Implementation Reference
- src/index.ts:146-152 (handler)MCP server request handler for the 'assess_pathogenicity' tool. Validates input using variantPredictionSchema, calls the AlphaGenomeClient.assessPathogenicity method, and formats the result as MCP content.case 'assess_pathogenicity': { const params = validateInput(variantPredictionSchema, args) as VariantPredictionParams; const result = await getClient().assessPathogenicity(params); return { content: [{ type: 'text', text: JSON.stringify(result, null, 2) }], }; }
- src/tools.ts:152-196 (schema)Tool schema definition including name, description, and input schema for validating parameters like chromosome, position, ref, alt, and optional tissue_type.export const ASSESS_PATHOGENICITY_TOOL: Tool = { name: 'assess_pathogenicity', description: `Comprehensive pathogenicity assessment of a genetic variant. Evaluates variant across all regulatory modalities and provides clinical classification. Returns: - Pathogenicity score (0-1 scale) - Clinical classification (pathogenic/likely_pathogenic/uncertain/likely_benign/benign) - Evidence breakdown (expression, splicing, TF binding impacts) Perfect for: clinical variant interpretation, pathogenicity prediction, diagnostic sequencing. Example: "Assess pathogenicity of chr19:44908684T>C"`, inputSchema: { type: 'object', properties: { chromosome: { type: 'string', description: 'Chromosome (chr1-chr22, chrX, chrY)', pattern: '^chr([1-9]|1[0-9]|2[0-2]|X|Y)$', }, position: { type: 'number', description: 'Genomic position (1-based)', minimum: 1, }, ref: { type: 'string', description: 'Reference allele', pattern: '^[ATGCatgc]+$', }, alt: { type: 'string', description: 'Alternate allele', pattern: '^[ATGCatgc]+$', }, tissue_type: { type: 'string', description: 'Optional: disease-relevant tissue (default: brain)', }, }, required: ['chromosome', 'position', 'ref', 'alt'], }, };
- src/tools.ts:709-730 (registration)Registration of the assess_pathogenicity tool by including ASSESS_PATHOGENICITY_TOOL in the ALL_TOOLS array, which is returned by the MCP listTools handler.export const ALL_TOOLS: Tool[] = [ PREDICT_VARIANT_TOOL, BATCH_SCORE_TOOL, ASSESS_PATHOGENICITY_TOOL, PREDICT_TISSUE_SPECIFIC_TOOL, COMPARE_VARIANTS_TOOL, PREDICT_SPLICE_IMPACT_TOOL, PREDICT_EXPRESSION_IMPACT_TOOL, ANALYZE_GWAS_LOCUS_TOOL, COMPARE_ALLELES_TOOL, BATCH_TISSUE_COMPARISON_TOOL, PREDICT_TF_BINDING_IMPACT_TOOL, PREDICT_CHROMATIN_IMPACT_TOOL, COMPARE_PROTECTIVE_RISK_TOOL, BATCH_PATHOGENICITY_FILTER_TOOL, COMPARE_VARIANTS_SAME_GENE_TOOL, PREDICT_ALLELE_SPECIFIC_EFFECTS_TOOL, ANNOTATE_REGULATORY_CONTEXT_TOOL, BATCH_MODALITY_SCREEN_TOOL, GENERATE_VARIANT_REPORT_TOOL, EXPLAIN_VARIANT_IMPACT_TOOL, ];
- src/alphagenome-client.ts:238-251 (helper)AlphaGenomeClient method that handles the assess_pathogenicity action by calling the Python bridge script with formatted parameters.async assessPathogenicity(params: VariantPredictionParams): Promise<any> { try { return await this.callPythonBridge('assess_pathogenicity', { chromosome: params.chromosome, position: params.position, ref: params.ref, alt: params.alt, tissue_type: params.tissue_type, }); } catch (error) { if (error instanceof ApiError) throw error; throw new ApiError(`Pathogenicity assessment failed: ${error}`, 500); } }
- scripts/alphagenome_bridge.py:255-296 (handler)Core implementation of pathogenicity assessment in the Python bridge. Computes a weighted pathogenicity score from expression, splicing, and TF binding impacts, and assigns ACMG-style clinical classification.def assess_pathogenicity(client, params: Dict[str, Any]) -> Dict[str, Any]: """ Comprehensive pathogenicity assessment of a variant. Uses all modalities to provide clinical interpretation. """ result = predict_variant_effect(client, params) # Calculate pathogenicity score based on multiple factors predictions = result['predictions'] # Score components expression_impact = abs(predictions.get('rna_seq', {}).get('fold_change', 0)) splice_impact = abs(predictions.get('splice', {}).get('delta', 0)) tf_impact = max([tf.get('change', 0) for tf in predictions.get('tf_binding', [])], default=0) # Combined pathogenicity score (0-1 scale) pathogenicity_score = min(1.0, (expression_impact * 0.4 + splice_impact * 0.4 + tf_impact * 0.2)) # Clinical classification if pathogenicity_score > 0.7: classification = 'pathogenic' elif pathogenicity_score > 0.4: classification = 'likely_pathogenic' elif pathogenicity_score > 0.2: classification = 'uncertain_significance' elif pathogenicity_score > 0.1: classification = 'likely_benign' else: classification = 'benign' return { 'variant': result['variant'], 'pathogenicity_score': float(pathogenicity_score), 'classification': classification, 'evidence': { 'expression_impact': float(expression_impact), 'splice_impact': float(splice_impact), 'tf_binding_impact': float(tf_impact) }, 'predictions': predictions }