search_genesLABSCO SUMMARY
Six databases behind one install, and nothing to sign up for.
A single gene question usually crosses three sites: NCBI for the gene, ClinVar for the variants, PubMed for the literature. Here they answer in one thread, which matters most when each answer decides what to look up next.
What it is
A Python MCP server over six public life-science sources: NCBI Gene, ClinVar, Taxonomy, sequence records, UniProt and PubMed. Ten tools, and no API key.
What you get
- Genes searched by symbol, name or function, and full detail by NCBI Gene ID
- Clinical variants and mutations from ClinVar
- DNA and protein sequences in FASTA or GenBank form
- Taxonomic classification for an organism
- Proteins searched by name, gene or organism, and detail by UniProt accession
- PubMed searched, an article's full abstract read, and the papers citing it listed
Requirements
Your client runs it with uvx --from bioinformatics-mcp-server bioinformatics-server. No API key — it reads the public NCBI and UniProt endpoints.
Setup effort
One command — uvx --from bioinformatics-mcp-server bioinformatics-server
USE CASES
01
Getting oriented on an unfamiliar gene
Search by symbol or function, then read the full record for the one that matches.
02
Checking what is known clinically about a variant
Query ClinVar for the variants and mutations recorded against the gene.
03
Pulling a sequence to work with
Fetch DNA or protein sequence in FASTA or GenBank, ready to paste into the next tool.
04
Going from gene to protein
Find the protein in UniProt and read its record by accession.
05
Catching up on the literature
Search PubMed and read the abstract in full rather than the title and a guess.
06
Following where a finding went
List the papers citing a publication, to see whether it held up.
10 tools
Listed in project docs. Your client may show a different list.
get_gene_detailssearch_clinvar_variantsget_taxonomyget_ncbi_sequencesearch_proteinsget_protein_detailssearch_pubmedget_pubmed_abstractget_pubmed_citationsBEFORE YOU INSTALL
What to know first
Tools you get
10 tools named in the project's documentation. None of them needs an API key; they cover NCBI, UniProt and PubMed.
Requires
uvx on PATH
INSTALLATION
Connect your client
Add the block to your client, or run the command directly.
Server setup
Configuration file
JSON
{
"mcpServers": {
"bioinformatics": {
"command": "uvx",
"args": [
"--from",
"bioinformatics-mcp-server",
"bioinformatics-server"
]
}
}
}uvx
TERMINAL
uvx --from bioinformatics-mcp-server bioinformatics-server
