Consultants' Network

When the Data Isn’t There: Drug Discovery from Phenotype to Molecule
Joaquín Ortiz Masllorens

The industry starts at the target. The patient starts at the symptom. Between the two sits a path, from clinical sign to gene, to variant, to protein, to binding site, to molecule, and almost nobody walks the whole of it.

This talk follows that path end to end and stops at the two points where the data runs out. At the front, moving from a patient’s clinical signs to the gene behind them, in diseases rare enough that the measurements a model would need were never taken. At the back, finding somewhere to bind on a target that shows no pocket in any available structure, because the site only opens while the protein is moving.

Both are problems where the conventional answer is that there is not enough data to proceed. The talk sets out what still works when that is true: where physics-based simulation reaches further than learned models, where it does not, and how to tell the difference before the money is spent, including the cases where the honest answer is that nothing will work.