mcpField
M
mosaic for drug target intelligence

mosaic for drug target intelligence

Local
github.com/sourabhnkProductivity0Updated 2w ago

About mosaic for drug target intelligence

mosaic-mcp provides preclinical drug-discovery intelligence tools for researchers and teams. It connects to a user-managed PostgreSQL database and covers targets, compounds, literature, structures, and clinical pipelines, with explicit data-coverage states.

Key features

  1. 01Build target profiles and compound lists
  2. 02Retrieve target-related papers and structures
  3. 03Review clinical pipelines and validation evidence
  4. 04Analyze pathways, networks, and resistance bypasses
  5. 05Label coverage state for each data area

Use cases

  1. 01Compile biological and drug evidence for target candidates
  2. 02Find compounds and papers associated with a target
  3. 03Assess clinical trials and validation precedent
  4. 04Explore synthetic-lethal and resistance-bypass directions

MCP support

The project is an MCP server that exposes drug-target intelligence tools to clients over stdio. The self-hosted package requires the user's own PostgreSQL database; a hosted service also offers remote connection options.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"mosaic": {
"command": "uvx",
"args": [
"mosaic-mcp"
],
"env": {
"DATABASE_URL": "<DATABASE_URL>",
"MOSAIC_API_KEY": "<MOSAIC_API_KEY>",
"MOSAIC_TIER": "<MOSAIC_TIER>"
}
}
}
}
Run commanduvx mosaic-mcp
Environment variables to set
DATABASE_URLSecretPostgreSQL/Neon connection string for YOUR Mosaic knowledge graph (must end /mosaic_db). Bring your own database — no shared instance is provided.
MOSAIC_API_KEYSecretAPI key unlocking the 7 Pro tools (optional; the 5 free tools work without it)
MOSAIC_TIEROverride tier for local dev: free (default), pro, or enterprise

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
pypimosaic-mcp2.2.1

Tools

11

Saved tool list · updated 2w ago

ParameterTypeRequiredDescription
paramsanyYes

Details

From the project README

mosaic-mcp

Pre-clinical drug-target intelligence as an MCP server: target profiles, compounds, literature, structure, clinical pipeline, validation precedent and whitespace analysis — 11 tools, 5 free and 6 Pro. Every count carries a coverage state, so "we did not look" never renders as zero.

Which one do you want?

This package is bring-your-own-database. It is the MCP tool layer only — it ships the queries, not the data. You point it at a PostgreSQL instance you control, with the Mosaic schema, and it serves the tools over it. There is no public read-only credential for Mosaic's knowledge graph.

If you want Mosaic's curated knowledge graph — 60 oncology targets, ~128K compounds linked to them, 57,896 papers and 9,062 clinical trials (counted 2026-10-07), refreshed monthly — use the hosted server. It needs no database and no local install:

  • Claude.ai connector (OAuth, Streamable HTTP): https://mcp.getmosaic.dev/mcp
  • Remote MCP with an API key (SSE): https://mcp.getmosaic.dev/sse
  • Sign in, API keys and on-demand dossiers for any human gene: https://getmosaic.dev

Quick start (self-hosted)

pip install mosaic-mcp
With Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "mosaic": {
      "command": "mosaic-mcp",
      "env": {
        "DATABASE_URL": "postgresql://USER:PASSWORD@YOUR-HOST:5432/mosaic_db?sslmode=require"
      }
    }
  }
}
With Claude Code
claude mcp add mosaic -- mosaic-mcp
Standalone
export DATABASE_URL="postgresql://..."   # your own Postgres
mosaic-mcp                               # stdio transport

uvx mosaic-mcp and python -m mosaic_mcp.server work too. This package is stdio-only: --transport sse exits with NotImplementedError, because remote transport is the hosted server above.

Tools

Free (5)
ToolWhat it returns
mosaic_get_target_profileThe target dossier: biology, approved and clinical drugs, disease associations ranked by Open Targets, validation evidence, organisations, scores — each axis with its coverage state
mosaic_get_target_compoundsCompounds with measured activity on the target; sort by potency (default) or by max clinical phase
mosaic_get_target_papersPapers about the target, matched by NCBI Gene ID (PubTator3) where your database has the annotations, up to the declared corpus cutoff
mosaic_get_target_structureAlphaFold structural snapshot and pockets, with a caution when a pocket is lined by low-confidence residues
mosaic_get_coverage_gridA coverage grid for any human gene symbol: identity (HGNC) and structure (AlphaFold DB) resolve live; the other axes are marked queued

Free calls return up to 10 compounds or papers; Pro up to 50.

Pro (6)
ToolWhat it returns
mosaic_clinical_pipelineClinical trials for compounds acting on the target (a ChEMBL mechanism link or pChEMBL ≥ 6)
mosaic_target_validationAssay precedent: what has been tried, in what model system
mosaic_pathway_contextThe pathways the target sits in
mosaic_target_networkThe target's 1-hop neighbourhood: compounds, diseases, pathways, organisations, interacting proteins
mosaic_synthetic_lethal_whitespacePartners functionally coupled to the target with little drug activity — assessed candidates and unassessed partners kept apart
mosaic_resistance_bypass_mapCandidate resistance-bypass and escape targets

mosaic_request_dossier and mosaic_get_dossier (on-demand dossiers) run only on the hosted server; they need its fetch and synthesis pipeline.

Configuration

VariableRequiredDescription
DATABASE_URLYesPostgreSQL connection string for your own instance; the path must name the database that holds the Mosaic schema.
MOSAIC_TIERNopro (or enterprise) enables the Pro tools in a self-hosted install.
MOSAIC_API_KEYNoIn stdio mode any value is treated as Pro; it is not checked locally. Keys are validated by the hosted server.

Changelog

Full history in CHANGELOG.md. Latest: 2.1.1 — the mosaic-mcp command starts again (it had failed with ImportError since 1.7.0). 2.1.0 capped mcp below 2, without which a fresh install died on import, and carried the fixes from a live audit of the hosted server.

License

Apache 2.0 for the MCP server tools. The hosted knowledge graph data requires a subscription for Pro-tier access.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.