A New Model of Drug Discovery

What role do you see for AI in accelerated drug discovery and development?
 

AI can provide new insights into the biological processes driving diseases, help identify new therapeutic targets, and guide the design molecules optimized to best inhibit those targets. The possibilities are endless. We have an AI-run robotics lab that helps us synthesize and test molecules. And by combining these new technologies with the disease and clinical expertise of big pharma, we’ll have a new model of drug discovery that – we think – will rapidly deliver new therapies to patients in need. 

How was AI used in Insilico Medicine’s collaboration with the University of Cambridge?
 

Michele Vendruscolo, Professor of Biophysics at the University of Cambridge and research lead on this project, combined his FuzDrop method with our AI target discovery engine, PandaOmics. Our system uses AI to sift through massive quantities of data (including omics data, clinical trials data, grants, and patents) to identify targets. The system then makes connections between biological processes and diseases that can be acted upon by drugs. 

For this study, Vendruscolo and his team performed a large-scale multi-omic study that quantified the relative impact of protein phase separation (PPS) in regulating various pathological processes associated with human disease. They prioritized candidates with high PandaOmics and FuzDrop scores, and generated a list of possible therapeutic targets for human diseases linked with PPS.

To validate the findings experimentally, the researchers focused on three predicted Alzheimer’s disease targets (MARCKS, CAMKK2 and p62) in two cell models of Alzheimer’s disease, which supported their potential as new therapeutic targets. 

What is the FuzDrop method?
 

The FuzDrop method is able to predict which proteins will undergo phase separation through sequence-based identifications of both droplet-promoting regions and of aggregation-promoting regions within droplets. Being able to identify proteins implicated in PPS is itself an important breakthrough as we know that the process underlies many diseases.

More broadly, how do you envision AI integrating with drug discovery and development in the future?
 

Pharma companies will continue to embrace the use of AI for drug research and development. And to supplement the growing interest, I feel it will be critical to provide these companies with full end-to-end capability – from target identification to drug design to clinical trial prediction – to create one seamless pipeline. I also believe robotics and quantum computing will perform an increasingly important role in tandem with AI to bring additional speed and capacity to drug development. 

We are quickly approaching the era of truly personalized medicine; AI will pave the future.