Louis
Louis is an MCP server + Slack bot that turns a genome-scale CD4⁺ T-cell CRISPRi Perturb-seq screen (Marson/Pritchard 2025) into a discovery-and-memory assistant living inside Claude — no separate app, no API key.
Core trick
Exposes a Perturb-seq screen through MCP tools, weighting sources by provenance and having the model stress-test its own leads against screen data before recommending them.
Description
Louis is an MCP server + Slack bot that turns a genome-scale CD4⁺ T-cell CRISPRi Perturb-seq screen (Marson/Pritchard 2025) into a discovery-and-memory assistant living inside Claude — no separate app, no API key. Named for Pasteur, it discovers understudied, druggable regulator handles wired to a disease's own GWAS risk genes, gated by CRISPRi knockdown QC and activation state; run blind as a positive control, the same ranking re-derives the Th17 masters STAT3/BATF/IRF4 — 1 of 77 regulator clusters clears significance (Crohn's q=0.025): disease-calibrated, not a hub.
Its sharpest move is honesty. Handed its own flagship RA lead DOT1L, Louis stress-tests it A→C — the wiring is a cross-disease artifact, and because DOT1L props up regulatory T cells an inhibitor likely worsens RA — then returns the survivor, HDAC7 for Th17-driven colitis (favorable, peer-reviewed direction), with the go/no-go experiment. It weighs every source by provenance (a preprint is never decisive), reviews itself, and the whole lab writes back to a shared memory.
Why it matters: the bottleneck a bench scientist named isn't analysis — it's trust. A discovery tool that kills its own darling, with receipts, is one you can hand the survivors.
README excerpt
<p align="center"> <img src="docs/louis.png" width="150" alt="Louis — a CD4+ T-cell target detective"> </p> # 🔎 Louis *A CD4+ T-cell target detective — named for Louis Pasteur (pronounced "Louie"; like Claude, a French name): **nothing trusted until it's verified.** Grounded in the Marson/Pritchard genome-scale CD4+ T-cell Perturb-seq screen.* **Ask Louis to discover novel, druggable CD4+ T-cell drug targets for an autoimmune disease — grounded in a real CRISPR screen, validated against the scientific web AND the field's live community signal, and remembered in a shared lab knowledge base.** An **MCP server**: no separate app, no API key, no metered credits. You talk to Claude the way you already do; this server grounds every answer in the **Marson/Pritchard genome-scale CD4+ T-cell Perturb-seq** (Zhu, Dann, …, Pritchard, Marson 2025; [preprint](https://www.biorxiv.org/content/10.64898/2025.12.23.696273v1) · [CZI Virtual Cells Platform](https://virtualcellmodels.cziscience.com/dataset/genome-scale-tcell-perturb-seq)) — disease enrichment, gene-network influence, CRISPRi knockdown QC, and activation-state dependence. *Built for the Built with Claude: Life Sciences hackathon (Builder track).* <p align="center"> <b><a href="https://claude.ai/code/artifact/b4a412b5-2eeb-4447-992c-0771fad5398c">🌐 Try Louis live</a> · <a href="https://www.loom.com/share/4e9ab966e3c848828d0ebbe9b33c7b47">🎥 The 3-minute demo</a> · <a href="#install--three-ways-to-use