Kim Branson on Building Groundbreaking AI/ML GSK Models in Drug Discovery
By Brittany Wade
May 24, 2022 | Dr. Kim Branson’s long-standing fascination with small molecule drug discovery and structural biology led to an established career in machine learning. After studying in Australia, Branson received post-doctoral training at Stanford and now works as the Senior Vice President and Global Head of Artificial Intelligence (AI) and Machine Learning (ML) at GlaxoSmithKline (GSK). In the latest episode of Bio-IT World’s Trends from the Trenches podcast, Branson and host, Stan Gloss, co-founder of BioTeam, discuss machine learning, human genetics, functional genomics, and how each predicts various aspects of health.
Machine learning is far from new, but recently, “an explosion of measurement technology has come to biology,” Branson says. As a result, the complexity with which the scientific community can measure conditions has grown, and pharmaceutical companies are bringing drugs to market faster. At GSK, their focus combines machine learning, human genetics, functional genomics, and medical science to pinpoint known drug targets and anticipate a drug’s likelihood of success on the market.
Functional genomics technology mimics pharmacological agents by modifying gene expression. Machine learning provides clarity from the data generated. There are also clinical implications: scientists apply algorithms to quantify marker expressions in response to a drug, predicting how patients are likely to respond.
AI and Machine Learning in Pharma
Branson explains why thoughtfulness is paramount to ML experiment design and why a project’s impact should be the primary consideration in AI. “You have to be very thoughtful about where you apply your machine learning,” says Branson. “Choosing the right problem will give us a step-change in innovation.” For example, distinguishing major causally related disease targets and their expression location while coupling them with clinical feedback yields unprecedented gains in the healthcare industry.
Machine learning can also characterize diseases and build “bridging models” to represent elements of a condition as a clinical translation system. Branson highlights his work in creating “biological twins”—3-D tumor culture models that replicate cancer and their respective immune environments. Branson and his team monitor tumor and immune responses to cancer drugs in real-time to anticipate a tumor’s every move within a patient.
AI and Ethics
When asked about AI and algorithm bias, Branson notes that GSK makes a concerted effort to minimize bias, write and design policy, and make informed policy suggestions as a transformative health technique. “At GSK, we make medicines for everybody,” says Branson. “At the same time, our AI approaches also need to apply to everybody.”
Trends from the Trenches Podcast
Bio-IT World’s Trends from the Trenches podcast delivers your insider’s look at the science, technology, and executive trends driving the life sciences through conversations with industry leaders. BioTeam co-founder Stan Gloss brings years of industry experience in science, data, and technology to conversations exploring what is driving data and discovery, and what’s coming next.
Catch up on our earlier episodes with GigaIO’s James Cuff, Alnylam’s James Bilotta, AWS’s Lita Sands, and NCI’s Tony Kerlavage, or subscribe.