Thermodynamics-Informed Neural Networks and Extensive Data Sets: Key Factors to Accurate Blind Predictions of Apparent pKa Values in the EuroSAMPL Challenge

Authors: Fraczkiewicz R
Publication: Physical Chemistry Chemical Physics
Keywords: admet predictor, machine learning (ML), pKa, pKa prediction
Software: ADMET Predictor®
Division: Cheminformatics

Abstract

Microscopic and macroscopic pKa values for 35 compounds selected by the organizers of euroSAMPL 1 challenge were blindly predicted with our thermodynamics-informed empirical S + pKa model (ranked submission 0x4cb7101f). Our results have received the first overall rank from the challenge organizers. We describe our methodology and discuss evaluation methods.

Graphical abstract: Thermodynamics-informed neural networks and extensive data sets: key factors to accurate blind predictions of apparent pKa values in the euroSAMPL challenge
By Robert Fraczkiewicz