Theoretical and Computational Chemistry

Artificial Neural Network to Predict Structure-based Protein-protein Free Energy of Binding from Rosetta-calculated Properties

Authors

  • Matheus Ferraz Aggeu Magalhães Institute, Oswaldo Cruz Foundation, FIOCRUZ, Recife, PE, 50670-465, Brazil & Department of Fundamental Chemistry, Federal University of Pernambuco, UFPE, Recife, PE, 50740-670, Brazil & Heidelberg Institute for Theoretical Studies,HITS, Heidelberg, 69118, Heidelberg, Germany ,
  • José Neto Recife Center for Advanced Studies and Systems, CESAR, Recife, PE, 50040-220, Brazil ,
  • Roberto Lins Aggeu Magalhães Institute, Oswaldo Cruz Foundation, FIOCRUZ, Recife, PE, 50670-465, Brazil & Department of Fundamental Chemistry, Federal University of Pernambuco, UFPE, Recife, PE, 50740-670, Brazil ,
  • Erico Teixeira Recife Center for Advanced Studies and Systems, CESAR, Recife, PE, 50040-220, Brazil

Abstract

The prediction of the free energy (ΔG) of binding for protein-protein complexes is of general scientific interest that allows a variety of applications in the fields of molecular and chemical biology, material sciences, and biotechnology. Despite its centrality in understanding protein association phenomena and protein engineering, the ΔG of binding is a daunting quantity to be obtained theoretically. In this work, we devise a novel Artificial Neural Network model to predict the ΔG of binding for a given three-dimensional structure of a protein-protein complex with Rosetta-calculated properties. Our model presents a root-mean-square error of 1.667 kcal/mol outperforming available state-of-art tools. Validation of the model for a variety of protein-protein complexes is showcased.

Content

Thumbnail image of ann_dg.pdf