pathway-report

Generate a comprehensive pathway deep-dive report for: $ARGUMENTS

Use the MCP tools available to you to gather data from all relevant sources, then synthesize a single structured report. Follow the steps below in order. If a step fails or returns no data, note the gap and continue — do not stop the report.

Data Gathering Steps

1. Pathway Identification

• •If the input looks like a Reactome stable ID (starts with R-), call reactome_get_pathway with pathway_id set to that ID, with include_participants: true and include_hierarchy: true.
• •Otherwise, call reactome_get_pathway with query set to the input to search for matching pathways. Present the top hits and pick the most relevant one, then do a direct lookup with include_participants: true and include_hierarchy: true.

2. Pathway Overview

From the Reactome data, extract:

• •Pathway name, stable ID, and species
• •Summation/description
• •Position in the pathway hierarchy (parent pathways)
• •Sub-events (child pathways and reactions) — list the first 10-15
• •Compartments (cellular locations)
• •Whether it is disease-associated

3. Key Member Genes

• •From the participants list, identify the gene/protein participants (filter for UniProt/ReferenceGeneProduct entries).
• •Select up to 10 key genes from the participant list.
• •Call batch_gene_summary with those gene symbols (taxon: human) to get brief summaries.
• •For the top 3-5 most important genes, call uniprot_get_protein to get function descriptions and GO terms.

4. Protein Interactions Within the Pathway

• •Pick the 2-3 most central genes from the pathway.
• •Call string_get_interactions for each to see how pathway members interact. Focus on interactions between pathway members rather than external interactions.

5. Clinical Relevance

• •For the top 3-5 key genes, call clinvar_search to find pathogenic/likely pathogenic variants.
• •Summarize the most clinically significant variants (up to 3 per gene, 10 total max).
• •Note which diseases are associated with variants in pathway members.

6. KEGG Cross-Reference

• •Call kegg_get_pathway with the pathway name or key gene to find the equivalent KEGG pathway (if any).
• •Note any additional member genes or connections found in KEGG but not Reactome.

7. Literature

• •Call pubmed_search with the pathway name to find recent relevant publications (limit: 10).
• •If few results, also search with the names of 2-3 key member genes combined with the pathway topic.

Report Format

Present the report in this structure:

Pathway: [Name] ([Reactome ID])

Summary: [1-2 sentence pathway description]

Hierarchy: [Top-level pathway] > [Parent pathway] > [This pathway]

Compartments: [cellular locations]

Disease-associated: Yes/No

Sub-pathways & Reactions

[Numbered list of child events, noting their type (Pathway vs Reaction)]

Key Member Genes ([N] total participants)

Protein Interaction Network

[Summary of how key pathway members interact, noting interaction scores and key hubs]

Clinical Variants in Pathway Members

Disease associations: [Summary of diseases linked to this pathway through variant data]

KEGG Cross-Reference

[KEGG pathway ID and name if found, plus any additional insights]

Recent Literature

Key Takeaways

[3-5 bullet points summarizing the most important findings: biological role, clinical significance, key genes, and any notable patterns]