Session 1: Artificial Intelligence & Cancer

Complex and simple Al/ML models for harnessing molecular data in oncology

With the growing availability of molecular and imaging data, in silico model engineering for creating computational replicas of biological processes in normal physiology and disease has become more important than ever. In this talk, I will first use the task of predicting drug sensitivity in cell lines as a context to showcase the strengths and limitations of existing computational models across various complexities and scales in the context of real-world oncology drug discovery. In the second part, I will explore the pivotal role of AI/ML modeling in early drug discovery, focusing on target identification. I will highlight the application of Graph Neural Networks and knowledge graphs for anticipating potential adverse effects associated with target modulation by therapeutic interventions.