AlphaGenome MCP Server

Instructions

Implementation Reference

• src/tools.ts:662-682 (schema)

  Tool schema definition with input validation for chromosome, position, ref, alt, and optional tissue_type.

  export const GENERATE_VARIANT_REPORT_TOOL: Tool = {
    name: 'generate_variant_report',
    description: `Generate comprehensive clinical report for a variant.
  
  Full analysis with all modalities and clinical interpretation.
  
  Perfect for: clinical reports, diagnostic summaries.
  
  Example: "Generate full report for chr13:32912345G>T"`,
    inputSchema: {
      type: 'object',
      properties: {
        chromosome: { type: 'string', pattern: '^chr([1-9]|1[0-9]|2[0-2]|X|Y)$' },
        position: { type: 'number', minimum: 1 },
        ref: { type: 'string', pattern: '^[ATGCatgc]+$' },
        alt: { type: 'string', pattern: '^[ATGCatgc]+$' },
        tissue_type: { type: 'string' },
      },
      required: ['chromosome', 'position', 'ref', 'alt'],
    },
  };

• src/tools.ts:709-730 (registration)

  Registration of all tools including GENERATE_VARIANT_REPORT_TOOL in the ALL_TOOLS array returned by ListTools endpoint.

  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/index.ts:265-271 (handler)

  MCP CallToolRequestSchema handler case that validates input and calls AlphaGenomeClient.generateVariantReport.

  case 'generate_variant_report': {
    const params = validateInput(variantPredictionSchema, args) as VariantPredictionParams;
    const result = await getClient().generateVariantReport(params);
    return {
      content: [{ type: 'text', text: JSON.stringify(result, null, 2) }],
    };
  }

• src/alphagenome-client.ts:448-455 (handler)

  AlphaGenomeClient method that proxies the tool call to Python bridge with action 'generate_variant_report'.

  async generateVariantReport(params: VariantPredictionParams): Promise<any> {
    try {
      return await this.callPythonBridge('generate_variant_report', params);
    } catch (error) {
      if (error instanceof ApiError) throw error;
      throw new ApiError(`Variant report generation failed: ${error}`, 500);
    }
  }

• scripts/alphagenome_bridge.py:658-676 (handler)

  Core implementation: generates variant report using assess_pathogenicity results, formats clinical classification, scores, evidence summary, and recommendations.

  def generate_variant_report(client, params: Dict[str, Any]) -> Dict[str, Any]:
      """Generate clinical report for variant."""
      result = assess_pathogenicity(client, params)
  
      # Build clinical report
      report = {
          'variant': result['variant'],
          'classification': result['classification'].upper(),
          'pathogenicity_score': result['pathogenicity_score'],
          'evidence_summary': {
              'expression_impact': f"{result['evidence']['expression_impact']:.4f}",
              'splicing_impact': f"{result['evidence']['splice_impact']:.4f}",
              'tf_binding_impact': f"{result['evidence']['tf_binding_impact']:.2f}"
          },
          'recommendation': 'Further clinical evaluation recommended' if result['pathogenicity_score'] > 0.5 else 'Routine monitoring',
          'generated_date': 'API call timestamp'
      }
  
      return report