Looks up proteins, structures, pathways, trials and papers from the primary databases, and runs the standard analyses on sequencing data.
Built for: The biologist or bioinformatician who spends the morning in six database tabs before the analysis can start.
The button opens the checkout, where the 4 servers and 6 skills are listed one by one. Every member is free, so the checkout lists them and asks for no card.
The reason to buy a stack rather than five listings: each line below needs more than one member connected at the same time.
Gather what is known about a target — UniProt entry, PDB structures, pathways and bioactive compounds — into one brief with the identifiers.
Find the trials that recruit for a condition and read their eligibility criteria and posted results.
Run an nf-core RNA-seq pipeline on your FASTQs and the single-cell QC that follows.
Search PubMed with MeSH terms for the papers behind a finding and follow their citations.
What each part contributes, and why it was picked over the obvious alternative. Prices and permissions are read from the listings, so nothing here can disagree with the catalogue.
The RCSB Protein Data Bank's own server: structure search by text, sequence, chemistry or motif, and entry details.
UniProt entries, sequences, features, variants, GO terms and cross-references, with release-aware provenance.
Searches ClinicalTrials.gov, reads study records and results, and matches eligibility criteria.
PubMed search with MeSH and PICO, related and citing articles, for the literature behind a result.
Runs nf-core pipelines (rnaseq, sarek, atacseq) on local FASTQs or public GEO/SRA data.
Quality control for single-cell RNA-seq with scverse practice: MAD-based filtering and the plots to justify it.
Integration, batch correction and multimodal analysis of single-cell data with scvi-tools.
Bioactivity data from ChEMBL — compounds, targets, IC50 and Ki — for structure-activity questions.
Pathway analysis and enrichment against Reactome.
RDKit for SMILES, descriptors, fingerprints, substructure and similarity search.
Every member is free, so the monthly figure is $0 whatever you do with the stack: nothing here is billed by the call, and only adding a paid member could change it.
| Part | What you are paying for | Monthly, as quoted |
|---|---|---|
| RCSB PDB | Free | — |
| UniProt | Free | — |
| Clinicaltrialsgov | Free | — |
| Pubmed Search | Free | — |
| Skills | ||
| nf-core Pipeline Deployment | Free · context cost only | — |
| Single-Cell RNA-seq Quality Control | Free · context cost only | — |
| scvi-tools Deep Learning | Free · context cost only | — |
| ChEMBL Database | Free · context cost only | — |
| Reactome Database | Free · context cost only | — |
| RDKit Cheminformatics Toolkit | Free · context cost only | — |
| Everything above | $0 of servers plus $0 of skills, the same in a quiet month and a busy one | $0/mo |
Subscriptions at their monthly plan price; one-off licences spread over 12 months.
Install only these and the bill is nothing: 4 servers and 6 skills, and all 83 tools.
Nothing in this stack carries a call allowance, so nothing here can run out before the month does. The bill is decided when you install it, not when you use it.
The spend cap is one figure for the whole account, set in your dashboard and checked when an order is placed. On the call path what stops a retry loop is the plan's call allowance and its rate limit, and a refused call is never billed.
The steps below come before the install lines: what to decide and scope before connecting anything, in the order it matters. The lines themselves follow, part by part, for the client you pick.
npx mcprush@latest stack add lab-desk writes all 4 servers into Claude Code’s config in one go — or run the lines below one by one. The 6 skills are not included — add them with lines 05–10.
# 01 · RCSB PDB
claude mcp add --scope user rcsb-rcsb-pdb-mcp '--' uvx rcsb-mcp
# 02 · UniProt
claude mcp add --scope user smaniches-uniprot-mcp '--' uvx uniprot-mcp-server
# 03 · Clinicaltrialsgov
claude mcp add --scope user clinicaltrialsgov-mcp '--' npx -y clinicaltrialsgov-mcp-server
# 04 · Pubmed Search
claude mcp add --scope user pubmed-search-mcp '--' uvx pubmed-search-mcp
# 05 · nf-core Pipeline Deployment
npx mcprush@latest skill add anthropics/nf-core-pipeline-deployment
# 06 · Single-Cell RNA-seq Quality Control
npx mcprush@latest skill add anthropics/single-cell-rna-seq-quality-control
# 07 · scvi-tools Deep Learning
npx mcprush@latest skill add anthropics/scvi-tools-deep-learning
# 08 · ChEMBL Database
npx mcprush@latest skill add k-dense-ai/chembl-database
# 09 · Reactome Database
npx mcprush@latest skill add k-dense-ai/reactome-database
# 10 · RDKit Cheminformatics Toolkit
npx mcprush@latest skill add k-dense-ai/rdkit-cheminformatics-toolkitnpx).uvx, docs.astral.sh/uv).Set-ExecutionPolicy -Scope CurrentUser RemoteSigned once, or type npx.cmd instead of npx (codex.cmd, gemini.cmd, grok.cmd likewise). PowerShell 7 needs neither.Nothing is connected until you approve it in Claude Code. mcprush is not in the path and meters nothing here.
Installed together, these tool surfaces add up. It is the first thing a security reviewer asks for, so what has been counted — and what nobody has counted yet — is on the page rather than in a PDF.
| Member | Tool surface | Write tools |
|---|---|---|
| Pubmed Search | imported | 5 |
| Clinicaltrialsgov | not imported | not recorded |
A stack's blast radius is the union of its members, not the worst of them. That union cannot be taken here, because 1 member of 4 has no imported tool surface — so the figure a review asks for is missing rather than low, and a member marked not imported is one nobody has counted rather than one that cannot write.
A stack is a default, not a verdict. These are the substitutions the maintainer would make, and what each one costs or saves.
Europe PMC adds preprints and full-text open-access articles to the PubMed record, for searches that need more than abstracts.
Written by the maintainer, kept on the page rather than in a support thread.
It does not give clinical advice. Trial eligibility here is a literature question, not a decision about a patient.
Pipelines run on your hardware or cluster. nf-core needs Nextflow and a container engine, and a whole-genome run needs the compute for it.
Database answers are as current as each database's release; the UniProt member records which release it read.
A stack has an owner: whoever keeps it re-checks the combination when a member changes, and the members themselves are published by the people named on each row.
Tools to inspect code changes, run checks and prepare a pull request review.
Tools to investigate a bug report and work toward a reproducible test case.
Tools to investigate alerts, check recent changes and follow the runbook.