Life science and bioinformatics databases your AI assistant can query

Answering a biological question often means hopping between protein, gene, variant, expression and biodiversity databases, each with its own identifiers and query language. Kahubi connects to UniProt, the Human Protein Atlas, ProteomicsDB, gnomAD, Open Targets, Pharos, the HuBMAP Human Reference Atlas, RCSB PDB structures, NCBI, PRIDE and MetaboLights, plus GBIF, iNaturalist and the Paleobiology Database for organisms and fossils.

The assistant translates your question into each database’s syntax, follows identifiers from one source to the next (gene to protein to structure), and cites accessions such as P38398, ENSG ids, rsIDs, PXD and MTBLS numbers so every claim can be checked.

15 sources you can connect

Questions researchers ask these sources

  • “Give me the UniProt entry, known structures and Reactome pathways for human TP53.”

    The UniProt accession, PDB IDs and Reactome pathway IDs, each verifiable at the source.

    TogoMCP
  • “Which targets have the strongest genetic evidence for Parkinson’s disease, and which already have drugs in development?”

    Association scores by evidence type, Ensembl target IDs and known drugs with their clinical phase.

    Open Targets Platform
  • “Find experimental structures of the SARS-CoV-2 main protease with a bound ligand below 2 Å resolution.”

    PDB IDs with method, resolution and ligand details.

    Protein structures
  • “How many pathogenic BRCA2 variants are listed in ClinVar?”

    The total count from an Entrez search, plus summaries of the top hits with clinical significance.

    NCBI E-utilities
  • “Where has the brown bear been recorded in Sweden since 2020?”

    Occurrence records with date, locality, coordinates, dataset and basis of record, plus the total count.

    GBIF
  • “Where has the invasive Spanish slug (Arion vulgaris) been observed in Sweden since 2020?”

    Research-grade observations with date, place, coordinates and observer.

    iNaturalist

Frequently asked questions

Do I need to know each database’s query syntax?
No. Ask the question as you would ask a colleague. The assistant writes the UniProt query, the gnomAD GraphQL or the SPARQL behind the scenes, and shows the identifiers it found.
Can it combine results across databases?
Yes. A typical flow is a gene in the Human Protein Atlas for tissue expression, the protein in UniProt for function and structures, and gnomAD for variant constraint, all in one answer with each fact cited to its source.
Are these databases free for commercial research?
Kahubi only lists sources whose licences allow use in a commercial product, and checks each one before adding it. Most are CC BY, CC0 or EMBL-EBI terms.

How MCP grounds AI answers in real records

Search life science sources in one research chat

Free to start. Connect a source in one click and get answers that cite the records they came from.