K: Protein-Ligand Absolute Binding Affinity Prediction via 3D-Convolutional Neural Networks.

Journal: Journal of chemical information and modeling
Published Date:

Abstract

Accurately predicting protein-ligand binding affinities is an important problem in computational chemistry since it can substantially accelerate drug discovery for virtual screening and lead optimization. We propose here a fast machine-learning approach for predicting binding affinities using state-of-the-art 3D-convolutional neural networks and compare this approach to other machine-learning and scoring methods using several diverse data sets. The results for the standard PDBbind (v.2016) core test-set are state-of-the-art with a Pearson's correlation coefficient of 0.82 and a RMSE of 1.27 in pK units between experimental and predicted affinity, but accuracy is still very sensitive to the specific protein used. K is made available via PlayMolecule.org for users to test easily their own protein-ligand complexes, with each prediction taking a fraction of a second. We believe that the speed, performance, and ease of use of K makes it already an attractive scoring function for modern computational chemistry pipelines.

Authors

  • José Jiménez
    Computational Biophysics Laboratory, Universitat Pompeu Fabra , Parc de Recerca Biomèdica de Barcelona, Carrer del Dr. Aiguader 88, Barcelona 08003, Spain.
  • Miha Škalič
    Computational Biophysics Laboratory, Universitat Pompeu Fabra , Parc de Recerca Biomèdica de Barcelona, Carrer del Dr. Aiguader 88, Barcelona 08003, Spain.
  • Gerard Martínez-Rosell
    Acellera , Carrer del Dr Trueta, 183 , 08005 Barcelona , Spain.
  • Gianni De Fabritiis
    Computational Science Laboratory , Parc de Recerca Biomèdica de Barcelona , Universitat Pompeu Fabra , C Dr Aiguader 88 , Barcelona , 08003 , Spain . Email: gianni.defabritiis@upf.edu.