What problem does it solve? Retrieving accurate protein function, sequence, taxonomy, and cross-reference data requires navigating UniProt's REST and SPARQL APIs, which involve complex query syntax, pagination, and dataset selection. This Skill provides a single Python wrapper that handles searching, streaming, counting, ID mapping, and historical lookups without manual API construction. ## Core Features & Use Cases - Protein Search & Retrieval: Search UniProtKB, UniParc, and UniRef with field-specific queries, or fetch a single entry by accession, including historical versions via UniSave. - Bulk Data Streaming: Stream up to 10 million entries in TSV or FASTA format for large-scale annotation pipelines, with automatic pagination for smaller searches. - ID Mapping & SPARQL: Convert identifiers between UniProt and 100+ external databases (PDB, Ensembl, RefSeq, GeneID), and run SPARQL queries for exact sequence matching and complex graph discovery. - Use Case: A researcher needs all reviewed human proteins with their gene names. They run a count to estimate volume, then stream taxonomy_id:9606 AND reviewed:true to a TSV file for downstream analysis. ## Quick Start Ask the AI to search UniProt for reviewed human p53 entries and show the first five results with their accessions and gene names.
