AI drug discovery
Bispecific Small Molecules

Most diseases are highly networked, meaning therapies often need to hit multiple nodes to have a sustainable effect. We have extended our AI platform so it can flexibly design and evolve compounds against more than one target.

The most accessible implementation is for dual targets and for these projects we refer to our dual-targeted compounds as bispecific small molecules (not to be confused with bispecific antibodies)

A range of projects are possible with this technology. At the simplest it could be the deliberate design of compounds to selectively hit two members of the same or closely related families where there is either predicted or known complementarity of binding sites.

At the other end of the scale the approach can be applied to explore whether completely unrelated targets might have binding sites that are sufficiently complementary to allow for the design of a single compound that can productively bind to both targets and have a meaningful therapeutic effect. Not all target pairs will be amenable, either because their sites are too dissimilar or because there is insufficient seed data for each individual target. For this reason we have scaled our algorithms at Exscientia so they can evolve compounds for thousands of dual target opportunities and identify those with the strongest chemical tractability and therapeutic potential.

This innovative approach is now well-validated, with bispecific small molecules successfully designed for both our internal portfolio as well as in collaborations with Sanofi and Dainippon Sumitomo.

We previously announced strong progress in our projects partnered with Dainippon Sumitomo for the treatment of psychiatric disease and more recently announced that Sanofi would exercise their option for an Exscientia-invented bispecific small molecule to treat immunological diseases through two unrelated targets.