David Schaller

This paper describes the development of KinoML, a framework for structure-enabled machine learning in drug discovery, focusing on kinases, and highlights key lessons in data management and model reproducibility.

The paper evaluates structure-based methods to predict how kinase mutations affect inhibitor binding, finding they accurately classify mutations as resistant or sensitizing.

This study benchmarks docking strategies for generating kinase:inhibitor complex geometries to improve machine learning scoring in drug discovery, finding that combining shape and electrostatics-based methods achieves the highest success rate.