Poisson-Boltzmann based machine learning (PBML) model for electrostatic analysis

Aug 4, 2026

Upload by Hazel

Electrostatics plays a central role across chemistry, physics, biology, and medicine, with the Poisson–Boltzmann (PB) framework as a key tool for electrostatic analysis. Accurate PB electrostatic solvation free energies for macromolecules remain difficult due to PB nonlinearity, dielectric jumps, charge singularities, and geometric complexity. This work proposes a Poisson–Boltzmann based machine learning (PBML) model trained with the second-order accurate MIBPB solver, achieving higher accuracy and faster inference than leading PB solvers. The model predicts PB solvation free energy for new biomolecules or MD-generated conformations with substantially reduced computational cost.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

Abstract

Significance

Introduction

Electrostatics in molecular science

Implicit vs explicit solvent models

Applications of PB models

PB solvers and numerical challenges

Matched interface and boundary (MIB) method and MIBPB

ESES for stability and robustness

Computational cost and need for transferability

arXiv :2312 . 11482v1 [physics .chem-ph] 29 Nov 2023 Manuscript submitted to BiophysicalJournal Article Poisson-Boltzmann based machine learning (PBML) model for electrostatic analysis Jiahui Chen1 , Yongjia Xu2 , Xin Yang3 , Zixuan Cang4 , Weihua Geng3,*, and Guo-Wei Wei5,6,* 1 Department of Mathematics, University of Arkansas, Fayetteville, AR 72701, USA 2 Google LLC, 1600 Amphitheater Pkwy, Mountain View, CA 94043, USA 3 Department of Mathematics, Southern Methodist University, Dallas, TX 75275, USA 4 Department of Mathematics, North Carolina State University, Raleigh, NC 27695, USA 5 Department of Mathematics, Michigan State University, MI 48824, USA 6 Department of Biochemistry and Molecular Biology, Michigan State University, MI 48824, USA * [Correspondence: wgeng@smu.edu](Correspondence: wgeng@smu.edu), [wei@math.msu.edu](wei@math.msu.edu) ABSTRACT Electrostatics is of paramount importance to chemistry, physics, biology, and medicine. The PoissonBoltzmann (PB) theory is a primary model for electrostatic analysis. However, it is highly challenging to compute accurate PB electrostatic solvation free energies for macromolecules due to the nonlinearity, dielectric jumps, charge singularity , and geometric complexity associated with the PB equation. The present work introduces a PB based machine learning (PBML) model for biomolecular electrostatic analysis. Trained with the second-order accurate MIBPB solver, the proposed PBML model is found to be more accurate and faster than several eminent PB solvers in electrostatic analysis. The proposed PBML model can provide highly accurate PB electrostatic solvation free energy of new biomolecules or new conformations generated by molecular dynamics with much reduced computational cost. SIGNIFICANCE This manuscript provides a Poisson-Boltzmann based machine learning (PBML) model for biomolecular electrostatic analysis. The features as the input to the ML models are generated with mathematical algorithms using biomolecular structures and force field. The learned model, which is trained using the most accurate PB solver MIBPB on more than 4000 biomolecules shows improved efficiency and accuracy in electrostatic analysis compared with the popular PB solvers. 1 INTRODUCTION Electrostatics is ubiquitous in the molecular world. The analysis of molecular electrostatics is of crucial importance to the bioscience research community. There are two significant types of electrostatic analyses, namely, qualitative analysis for general electrostatic characteristics, such as visualization and electrostatic steering, and quantitative analysis for statistical, thermodynamic and/or kinetic observable, such as solvation free energy, solubility, and partition coefficient. Molecular electrostatics can be analyzed by explicit or implicit models. Explicit solvent models resolve electrostatic effect in atomic detail and thus are more accurate but can be very expensive for large biomolecular systems. Implicit solvent models describe the solvent as a dielectric continuum, while the solute molecule is modeled with an atomistic description (1) . A wide variety of two-scale implicit solvent models has been developed for electrostatic analysis, including generalized Born (GB) (2), polarizable continuum (3) and Poisson-Boltzmann (PB) models (4) . PB models have been applied to calculating protein titration states (5), protein-protein and protein-ligand binding energetics (6), RNA nucleotide protonation(7), chromatin packing (8), etc. The PB theory has also been used for the evaluation of biomolecular electrostatic forces for molecular Langevin dynamics or Brownian dynamics (9) . GB methods are faster than PB methods, but provide only heuristic estimates for PB electrostatic energies. Due to its success in describing biomolecular systems, the PB model has attracted a wide attention in both mathematical and biophysical communities. In the past two decades, many efforts have been g

FAQs

Why is accurate Poisson–Boltzmann (PB) electrostatic solvation free energy computation challenging?

PB electrostatics is hard to compute accurately for macromolecules because the PB equation is nonlinear and involves dielectric jumps, charge singularities, and difficult biomolecular geometry at continuum–discrete interfaces.

What is the PBML model proposed in this work?

The PBML model uses features derived from biomolecular structures and force-field information as inputs to machine learning models, with training guided by accurate PB computations from the second-order MIBPB solver.

How does PBML improve performance compared with existing PB solvers?

Trained using the most accurate PB solver MIBPB on over 4000 biomolecules, the PBML model yields improved efficiency and accuracy, providing highly accurate PB electrostatic solvation free energy for new biomolecules or MD-generated conformations with much lower computational cost.

Aug 4, 2026

Upload by Hazel