Safe and effective medications are needed to meet the medical needs of billions worldwide, which are driven by aging populations and increasing insight into disease burden. However, getting a novel drug to the market currently takes 13–15 years and US$2–3 billion, on average1. Faced with skyrocketing costs and high failure rates, researchers are looking at ways to make drug discovery and development more efficient through automation, artificial intelligence (AI) and new data modalities2,3. AI has become woven into therapeutic discovery since the emergence of deep learning4. It stands out as an approach to guide discovery5 by finding and extracting actionable predictions that lend themselves to hypotheses testable in the laboratory.