Stacking Gaussian processes to improve $$pK_a$$ predictions in the SAMPL7 challenge

Raddi, Robert; Voelz, Vincent A.

doi:10.1007/s10822-021-00411-8

Cited by 3 publications

(1 citation statement)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The atomic charge describes the local electrostatic environment around the ionization sites and reflects the local electronic structure polarization. Thus, it is a common choice for p K a -related tasks in the feature sets of traditional statistical learning methods, − ,− a part of molecular embeddings in deep learning methods, − and the semiempirical correction of quantum chemistry in solvent models. ,, In addition to the formal charge, the performance of Gasteiger’s empirical charge scheme and GFN2-xTB partial charge is also explored during the whole training and exploration process. We choose the formal charge as default because it is directly read from the SMILES input without any further time-consuming calculation and achieves the best accuracy on the most diverse Novartis, SAMPL6, and SAMPL8 data sets (Table S5) among the three charge schemes.…”

Section: Discussionmentioning

confidence: 99%

Bridging Machine Learning and Thermodynamics for Accurate pK_a Prediction

Luo,

Zhou,

Zhu

et al. 2024

JACS Au

View full text Add to dashboard Cite

Integrating scientific principles into machine learning models to enhance their predictive performance and generalizability is a central challenge in the development of AI for Science. Herein, we introduce Uni-pK a , a novel framework that successfully incorporates thermodynamic principles into machine learning modeling, achieving high-precision predictions of acid dissociation constants (pK a ), a crucial task in the rational design of drugs and catalysts, as well as a modeling challenge in computational physical chemistry for small organic molecules. Uni-pK a utilizes a comprehensive free energy model to represent molecular protonation equilibria accurately. It features a structure enumerator that reconstructs molecular configurations from pK a data, coupled with a neural network that functions as a free energy predictor, ensuring high-throughput, data-driven prediction while preserving thermodynamic consistency. Employing a pretrainingfinetuning strategy with both predicted and experimental pK a data, Uni-pK a not only achieves state-of-the-art accuracy in chemoinformatics but also shows comparable precision to quantum mechanics-based methods.

show abstract