skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Binding Affinity Prediction by Pairwise Function Based on Neural Network

Abstract

In this paper, we present a new approach to estimate the binding affinity from given three-dimensional poses of protein–ligand complexes. In this scheme, every protein–ligand atom pair makes an additive free-energy contribution. The sum of these pairwise contributions then gives the total binding free energy or the logarithm of the dissociation constant. The pairwise contribution is calculated by a function implemented via a neural network that takes the properties of the two atoms and their distance as input. The pairwise function is trained using a portion of the PDBbind 2018 data set. The model achieves good accuracy for affinity predictions when evaluated with PDBbind 2018 and with the CASF-2016 benchmark, comparing favorably to many scoring functions such as that of AutoDock Vina. The framework here may be extended to incorporate other factors to further improve its accuracy and power.

Authors:
ORCiD logo [1];  [1];  [1];  [1];  [1]
  1. Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Publication Date:
Research Org.:
Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA); American Heart Association (AHA)
OSTI Identifier:
1756151
Report Number(s):
LLNL-JRNL-800142
Journal ID: ISSN 1549-9596; 1002405
Grant/Contract Number:  
AC52-07NA27344
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
Journal of Chemical Information and Modeling
Additional Journal Information:
Journal Volume: 60; Journal Issue: 6; Journal ID: ISSN 1549-9596
Publisher:
American Chemical Society
Country of Publication:
United States
Language:
English
Subject:
59 BASIC BIOLOGICAL SCIENCES; Ligands; Crystal structure; Protein structure; Neural networks; Screening assays

Citation Formats

Zhu, Fangqiang, Zhang, Xiaohua, Allen, Jonathan E., Jones, Derek, and Lightstone, Felice C. Binding Affinity Prediction by Pairwise Function Based on Neural Network. United States: N. p., 2020. Web. doi:10.1021/acs.jcim.0c00026.
Zhu, Fangqiang, Zhang, Xiaohua, Allen, Jonathan E., Jones, Derek, & Lightstone, Felice C. Binding Affinity Prediction by Pairwise Function Based on Neural Network. United States. https://doi.org/10.1021/acs.jcim.0c00026
Zhu, Fangqiang, Zhang, Xiaohua, Allen, Jonathan E., Jones, Derek, and Lightstone, Felice C. 2020. "Binding Affinity Prediction by Pairwise Function Based on Neural Network". United States. https://doi.org/10.1021/acs.jcim.0c00026. https://www.osti.gov/servlets/purl/1756151.
@article{osti_1756151,
title = {Binding Affinity Prediction by Pairwise Function Based on Neural Network},
author = {Zhu, Fangqiang and Zhang, Xiaohua and Allen, Jonathan E. and Jones, Derek and Lightstone, Felice C.},
abstractNote = {In this paper, we present a new approach to estimate the binding affinity from given three-dimensional poses of protein–ligand complexes. In this scheme, every protein–ligand atom pair makes an additive free-energy contribution. The sum of these pairwise contributions then gives the total binding free energy or the logarithm of the dissociation constant. The pairwise contribution is calculated by a function implemented via a neural network that takes the properties of the two atoms and their distance as input. The pairwise function is trained using a portion of the PDBbind 2018 data set. The model achieves good accuracy for affinity predictions when evaluated with PDBbind 2018 and with the CASF-2016 benchmark, comparing favorably to many scoring functions such as that of AutoDock Vina. The framework here may be extended to incorporate other factors to further improve its accuracy and power.},
doi = {10.1021/acs.jcim.0c00026},
url = {https://www.osti.gov/biblio/1756151}, journal = {Journal of Chemical Information and Modeling},
issn = {1549-9596},
number = 6,
volume = 60,
place = {United States},
year = {Mon Apr 27 00:00:00 EDT 2020},
month = {Mon Apr 27 00:00:00 EDT 2020}
}

Works referenced in this record:

Thermodynamics of Protein–Ligand Interactions: History, Presence, and Future Aspects
journal, January 2004


The Many Roles of Computation in Drug Discovery
journal, March 2004


Recent developments in structure-based drug design
journal, February 2000


Advancing Drug Discovery through Enhanced Free Energy Calculations
journal, July 2017


Assessment of programs for ligand binding affinity prediction
journal, January 2008


Classification of Current Scoring Functions
journal, February 2015


Protein–Ligand Scoring with Convolutional Neural Networks
journal, April 2017


PotentialNet for Molecular Property Prediction
journal, November 2018


Comparative Assessment of Scoring Functions on a Diverse Test Set
journal, April 2009


Comparative Assessment of Scoring Functions: The CASF-2016 Update
journal, November 2018


Critical assessment of the automated AutoDock as a new docking tool for virtual screening
journal, November 2006


The PDBbind Database:  Collection of Binding Affinities for Protein−Ligand Complexes with Known Three-Dimensional Structures
journal, June 2004


Forging the Basis for Developing Protein–Ligand Interaction Scoring Functions
journal, January 2017


The Amber biomolecular simulation programs
journal, January 2005


ff14SB: Improving the Accuracy of Protein Side Chain and Backbone Parameters from ff99SB
journal, July 2015


Development and testing of a general amber force field
journal, January 2004


Fast, efficient generation of high-quality atomic charges. AM1-BCC model: I. Method
journal, January 2000


Message passing interface and multithreading hybrid for parallel molecular docking of large databases on petascale high performance computing machines
journal, January 2013


In Need of Bias Control: Evaluating Chemical Data for Machine Learning in Structure-Based Virtual Screening
journal, October 2018


Extended-Connectivity Fingerprints
journal, April 2010


Development and validation of a genetic algorithm for flexible docking 1 1Edited by F. E. Cohen
journal, April 1997


General and targeted statistical potentials for protein-ligand interactions
journal, August 2005


Improving scoring-docking-screening powers of protein-ligand scoring functions using random forest
journal, November 2016


Machine Learning in Drug Discovery
journal, August 2018