What problem does it solve? Protein sequence analysis often requires querying multiple signature databases (Pfam, CDD, SMART, and more) individually, which is slow and redundant. This Skill provides unified programmatic access to InterPro, which integrates 14 member databases into one searchable resource, with built-in rate limiting, pagination, and retry handling. ## Core Features & Use Cases - Domain and Family Identification: Find all domains, families, repeats, and sites annotated on a protein, or find all proteins containing a given domain. - Domain Architecture (IDA) Search: Search for proteins by exact domain combinations and ordering using the two-step ida_search workflow. - InterPro-N Deep Learning Fallback: When standard signatures return no hits, query InterPro-N predictions to annotate uncharacterized or divergent proteins. - Use Case: Given a UniProt accession like P04637, fetch all InterPro entries, linked PDB structures, and taxonomy distribution for its domains, writing results to JSONL files for downstream pipeline processing. ## Quick Start Use the interpro-database skill to list all InterPro domains annotated on UniProt protein P04637 and save the results to a file.
