Drug discovery
Molecular OptimizationMaturity: concept
Conformation, energy, binding and molecular-property optimisation.
concept — An idea we find credible. Nothing has been built or measured.
What this is
The problem
Lead optimisation is multi-objective. A candidate must bind well, dissolve, survive metabolism, avoid toxicity and be synthesisable. Improving one property routinely damages another.
Where the current approach strains
Chemical space is discrete and vast. Medicinal chemists navigate it with expertise and intuition, and the search is inherently serial — make, test, learn, repeat.
What we are exploring
Multi-objective optimisation over molecular structure with synthesisability as a first-class constraint rather than an afterthought, since a molecule nobody can make is not a lead.
What would have to be true
Proposed molecules that a chemist judges both novel and makeable. That is a human evaluation, and it is the correct first gate — a benchmark score would not substitute.
Where it applies
Related
Accelerated Drug Screening
Virtual screening has to be cheap enough to run across a whole library and accurate enough not to throw away good molecules. Docking scores track binding affinity loosely, so shortlists carry false positives you can measure and false negatives nobody ever finds out about. We are working on scoring for the middle of the funnel, tested by whether it ranks known binders highly without being told.
Targeted Drug Design
Binding affinity, selectivity and off-target risk concepts for targeted compounds.
Targeting and Prediction
Protein/biological target identification and interaction prediction.