Artificial Intelligence (AI) has the potential to revolutionize how we find, develop, and deliver new life-saving therapies.
Join “State of the Art” host and Sidley partner Stephen Abreu as he speaks with Colin Hill, CEO and Co-Founder of Aitia. Together, they discuss the challenges and opportunities AI presents for the pharmaceutical industry, particularly how AI is accelerating the pace of drug discovery and its implications for the future of medicine.
Stephen and Colin discuss how biology became “quantitative” (2:30), initial efforts to map “genetic circuitry” (3:17), initial company forays into “computational biology,” “biosimulation” and the use of AI in biology (5:04), the history of creation of AI tools for use in biology and drug discovery (7:22), natural language model AI versus causal AI (21:52), “State of the Art” drug discovery AI models, companies that use them, and three grand problems (23:05), as well as Aitia’s “Digital Twins” technology (26:00).
Below are two excerpts from the discussion:
Stephen and Colin delve into a crucial facet of AI in drug discovery: the distinction between druggable and non-druggable targets. AI is reshaping the way we approach these targets, from identifying potential therapies for known diseases to uncovering novel solutions for previously uncharted territory.
“…I think there’s also a deeper and more interesting way of thinking about how you approach undruggable targets, which is, if you actually understood the blueprint, right, in terms of the connectivity contained in the models, then you could potentially find a different target that is more druggable and that may be upstream or downstream of that intended first target.”
– Colin Hill
Colin explains how multi-data-point “jigsaw puzzle pieces” of critical biological information are put together to create AI models of biological systems to potentially revolutionize the field of genomics.
“It’s really almost multivariate regression on steroids. Let’s just even make it simple. You have the expression levels of two genes… So, just as if you’re doing a regression plot, you’re looking at Gene A and Gene B, you’d have a diagonal line if they’re perfectly correlated, and they could be perfectly correlated because one is driving the other.”
– Colin Hill
This post is as of the posting date stated above. Sidley Austin LLP assumes no duty to update this post or post about any subsequent developments having a bearing on this post.