Off-target action is a major challenge in novel drug design -- molecules that bind with the target protein are also likely to bind with other similar proteins and introduce unintended side effects. The objective in this project is to create generative generative models that automatically avoid this off-target effect. This model will be combined with a pipeline of interpretable analysis to help medicinal chemists understand structural features that reduce off target binding. The project has been co-developed with Oxford Drug Design, who will be the industrial partner and advisor on the project.
Generative AI is major topic in drug and protein design. In this project, we will stude a whole range of such models (e.g. protein language models, diffusion and flow matching models etc) and create new versions that void off target binding.