Daphne Koller: AI-Driven Drug Discovery Using Digital Biology

Daphne Koller discusses how Insitro is using machine learning models to predict the outcome of drug development experiments and design novel, safe, and effective therapies.

• Drug development is becoming more challenging and expensive due to the high failure rate of experiments.
• Insitro is using high-quality data and machine learning models to predict the outcome of experiments and design novel therapies.
• The focus is on learning meaningful representations of clinical state using self-supervised machine learning models.
• Insitro has partnered with pharmaceutical companies to access data on liver disease and used machine learning to predict patient progression.
• The ultimate goal is to develop a new approach to drug development that helps more people faster and at a lower cost.

Insitro used a self-supervised machine learning model to analyze histopathology data from liver biopsies of patients with non-alcoholic steatohepatitis (NASH). The model was able to predict the pathologist's scores and other clinical covariates with a strong correlation, indicating that the variation was biologically meaningful. Patients who progressed over the course of the trial had a different machine learn score than those who did not progress, demonstrating the potential of machine learning to predict patient outcomes.

Modern medicine has provided effective tools to treat some of humanity’s most significant and burdensome diseases. At the same time, it is becoming consistently more challenging and more expensive to develop new therapeutics. The drug development process involves multiple steps, each of which requires a complex and protracted experiment that often fails. Insitro CEO and Founder Daphne Koller believes that, for many of these phases, machine learning models can help predict the outcome of the experiments and that those models, while inevitably imperfect, can outperform predictions based on traditional heuristics.In this keynote, Koller discusses how Insitro is bringing together high-quality data from human cohorts, while also developing cutting-edge methods in high-throughput biology and chemistry that can produce massive amounts of in vitro data relevant to human disease and therapeutic interventions.Koller also covers how the data is then used to train machine learning models that make predictions about novel targets, coherent patient segments, and the clinical effect of molecules. Insitro’s ultimate goal is to develop a new approach to drug development that uses high-quality data and ML models to design novel, safe, and effective therapies that help more people faster, and at a lower cost.Koller also co-founded Engageli and has served as the co-CEO and President of Coursera. She received the MacArthur Foundation Fellowship in 2004 and the ACM Prize in Computing in 2008, and was elected a fellow of the American Association for Artificial Intelligence in 2004, in addition to receiving many other honors and awards.