GSK tested protein candidates generated by Chai Discovery models and is gaining access to the models for drug discovery

GSK tested protein candidates generated by Chai Discovery models and is gaining access to the models for drug discovery

On October 9, Chai Discovery and GSK announced a collaboration. Before the agreement, GSK tested protein candidates in its own laboratory that Chai models had proposed for GSK’s therapeutic targets.

The search for a protein drug begins with choosing a therapeutic target: a molecule in the body whose activity researchers want to influence. They then look for a binder, a protein that can attach to that target. Chai proposed such candidates for GSK’s targets without training the model specifically for each target.

The model works before the experiment by proposing protein sequences. The laboratory tests the candidates against real molecules to measure whether they bind to the intended target and how strongly they bind. This provides experimental results for a computationally proposed candidate.

GSK tested these candidates against several targets and found binders for every target tested.

Following this evaluation, GSK and Chai expanded their work into a collaboration. GSK will gain access to Chai models to predict how proteins fold, which determines their shape, and to design new protein candidates for its drug discovery programs.