An open engine that ranks every drug pair by predicted synergy before anyone runs a trial
Use existing cell-line and organoid data to score thousands of drug pairs, publish the ranking openly, and only test the top of the list in people.
DepMap dependency screens, the NCI ALMANAC pairwise matrix and published organoid drug-response sets contain far more combination signal than has been mined. A public model that predicts synergy and, critically, therapeutic window (tumour versus normal-cell toxicity) for each pair in each molecular context would give trialists a prioritised shortlist. Models should be scored prospectively against every new combination readout.
- Too many combinations to test · There are thousands of possible drug pairs and sequences. Trials can test a few dozen a year.
- Preclinical models that do not predict people · Nine in ten cancer drugs that work in mice fail in humans. Our models are the reason.
Pages like this
not linked directly; found by shared links- IdeaA drug screen that only rewards killing sleeping cancer cells
Shares DrugBank & ChEMBL, DepMap (Cancer Dependency Map), CRISPR functional genomics, Patient-derived organoids.
- IdeaAutomated combination discovery: patient-sample screens feeding Bayesian platform trials
Shares Patient-derived organoids to pick ADC payloads, AI-driven drug & target discovery, Patient-derived organoids, Too many combinations to test.
- IdeaAn open foundation model of the cancer cell trained on perturbation data
Shares DepMap (Cancer Dependency Map), AI-driven drug & target discovery, CRISPR functional genomics, Too many combinations to test.
- IdeaMake in vivo metastasis screens a required step in drug discovery
Shares DepMap (Cancer Dependency Map), CRISPR functional genomics, Patient-derived organoids, Preclinical models that do not predict people.
- IdeaSelf-driving laboratories that run the cancer biology hypothesis loop autonomously
Shares AI-driven drug & target discovery, CRISPR functional genomics, Patient-derived organoids, Preclinical models that do not predict people.
- IdeaA virtual cancer cell that predicts what a drug will do before you test it
Shares AI-driven drug & target discovery, CRISPR functional genomics, Too many combinations to test, Preclinical models that do not predict people.
- IdeaShared reference organoid and PDX panels that every lab can test against
Shares A public atlas of drug-pair responses across a thousand patient-derived organoids, Patient-derived organoids, Preclinical models that do not predict people.
- IdeaGrow each trial patient's tumour as organoids to decide which platform arm opens next
Shares Patient-derived organoids to pick ADC payloads, Patient-derived organoids, Too many combinations to test, Preclinical models that do not predict people.