About this source
Federated structure search across the Protein Data Bank and computed models (AlphaFold DB, 3D-Beacons): free-text, sequence and organism/method/resolution filters. Sequence similarity via mmseqs2, fold similarity via Foldseek, structural alignment (TM-align), ligand and binding-site discovery, PDB-wide statistics, and UniProt/InterPro annotations for a protein. All upstream sources are keyless.
Questions you can ask Protein structures
Type them into Kahubi’s chat once the source is connected.
“Find experimental structures of the SARS-CoV-2 main protease with a bound ligand below 2 Å resolution.”
What comes back: PDB IDs with method, resolution and ligand details.
“Which structures have a fold similar to this AlphaFold model?”
What comes back: Foldseek hits with identifiers and similarity scores.
What the assistant can do with it
The 7 tools this server offers to Kahubi. The assistant picks the right one for each question.
protein_search_structuresSearch experimental (PDB) and predicted (computed-model) protein structures by free text, protein sequence (triggers an mmseqs2 similarity search), and/or organism, method, and...
protein_get_structureFetch structures with metadata and coordinate-file URLs.
protein_find_similarFind structurally or evolutionarily related proteins.
protein_track_ligandsLigand discovery and binding-site analysis across the PDB.
protein_compare_structuresStructurally align multiple structures (up to the configured batch cap) via the RCSB Structural Comparison service (TM-align / jFATCAT).
protein_analyze_collectionProfile the PDB into distributions and trends over an optional scoping query: counts by method, organism, or polymer composition; resolution and molecular-weight histograms; rel...
protein_get_annotationsSequence and functional annotation for a protein: UniProt features (domains, binding sites, PTMs), natural variants, and InterPro domain/family memberships (Pfam, PROSITE, …) wi...
Good to know
- Community server. An open-source server maintained by the MCP community. Kahubi tests it with live queries before listing it and health-checks it regularly.
- Beta. The server works, but answers can be slower and its tools may still change.
- Cited, checkable answers. Results are treated as retrieved material and cited with the source’s own identifiers, so you can open every record the answer relied on.
- Your files stay private. Only the search the assistant writes for your question is sent to Protein structures.
More life science sources
- TogoMCP (UniProt, ChEMBL, PubChem, PDB, Reactome, NCBI)One service across 30+ life-science databases: proteins, genes, compounds, structures, pathways, variants and MeSH.
- Open Targets PlatformTarget–disease–drug associations with the evidence behind them, from EMBL-EBI.
- NCBI E-utilitiesGene, ClinVar, dbSNP, Taxonomy, Nucleotide, Protein, PMC and other NCBI databases.
- GBIF (biodiversity occurrences)2.5 billion species occurrence records from museums, surveys and citizen science worldwide.
- iNaturalistCitizen-science observations with photos, research-grade identifications and places.
- Paleobiology DatabaseFossil occurrences, taxa, collections and diversity through geological time.