We report our recent work for the prediction of protein binding affinity using convolutional neural networks (CNNs). This work utilizes uniform features extraction from both topological data analysis and electrostatics from charged protein-ligand complexes. Our methods overcome the difficulties of involving electrostatics, which is expensive to compute due to its long range and pairwise natures. The simulation results show accurate prediction of binding affinity and meanwhile demonstrate the significance of involving electrostatics in the machine learning framework.
The role of electrostatics in machine learning models for biological applications
Elyssa Sliheet, Southern Methodist UniversityAuthors: Elyssa Sliheet
2023 AWM Research Symposium
Pure and Applied Talks by Mathematicians Enhancing Diversity in Graduate Education (EDGE) [Organized by Quiyana M. Murphy and Sofía Martínez Alberga]