KaMLs for Predicting Protein pK _a Values and Ionization States: Are Trees All You Need?

ABSTRACT: Despite its importance in understanding biology and computer-aided drug discovery, the accurate prediction of protein ionization states remains a formidable challenge. Physics-based approaches struggle to capture the small, competing contributions in the complex protein environment, while machine learning (ML) is hampered by scarcity of experimental data. Here we report the development of pK _a ML (KaML) models based on decision trees and graph attention networks (GAT), exploiting physicochemical understanding and a new experiment pK _a database (PKAD-3) enriched with highly shifted pK _a's. KaML-CBtree significantly outperforms the current state of the art in predicting pK _a values and ionization states across all six ti

SUBMITTER: Shen M 

PROVIDER: S-EPMC11601431 | biostudies-literature | 2025 Jan

REPOSITORIES: biostudies-literature