Skip to content
02 / biotech
Biotech

The pipeline before the molecule exists.

Targets, validation, and pharmacogenomic stratification across a curated omics database: 108K target-disease associations, 56 pharmacogenes, 135K resistance variants (shipping as an open SQLite bundle in a future release). Find a target, validate the evidence, and stratify the patients a program will treat and the patients it will spare.

target_discoveryvalidate_targetstratify_patientssearch_chemblsearch_literaturesearch_clinical_trials
One instruction

What it looks like in practice.

any MCP-compatible assistant
you Find targets for non-small-cell lung cancer and validate the strongest evidence.
ranked 38 targets via target_discovery over 108K associations
cross-checked literature + trials with validate_target
surfaced 1,284 resistance variants · 137 in the binding site
stratified responders with stratify_patients across 56 pharmacogenes
Benchmarked

The accuracy, stated plainly.

108K
target-disease
associations
56
pharmacogenes
indexed
135K
resistance
variants
13K
binding-site
variants

Omics-driven target discovery

A curated layer of target-disease associations, queried by disease. Ranked, traceable to source, and cited in every response.

Validation against the literature

Targets are cross-checked against papers, patents, and clinical trials, so a candidate target arrives with its supporting and contradicting evidence attached.

Resistance, mapped to structure

135K ClinVar pathogenic variants, with the subset that affects the binding site flagged. Resistance is a structural fact, mapped from the first stage.

Pharmacogenomic stratification

Patient stratification across 56 pharmacogenes. The responders and the non-responders a program identifies before it commits to a trial design.

Research preview

Start the pipeline at the target, with the evidence attached.

NovoMCP is open to a small group of PIs, postdocs, and research engineers. Tell us what you are working on.