Learning Protein-Ligand Binding Affinity with Atomic Environment Vectors

Abstract

<div> <div> <div> <p>Scoring functions for the prediction of protein-ligand binding affinity have seen renewed interest in recent years when novel machine learning and deep learning methods started to consistently outperform classical scoring functions. Here we explore the use of atomic environment vectors (AEVs) and feed-forward neural networks, the building blocks of several neural network potentials, for the prediction of protein-ligand binding affinity. The AEV-based scoring function, which we term AEScore, is shown to perform as well or better than other state-of-the-art scoring functions on binding affinity prediction, with an RMSE of 1.22 pK units and a Pearson’s correlation coefficient of 0.83 for the CASF-2016 benchmark. However, AEScore does not perform as well in docking and virtual screening tasks. We therefore show that the model can be combined with the classical scoring function AutoDock Vina in the context of ∆-learning, where corrections to the AutoDock Vina scoring function are learned instead of the protein-ligand binding affinity itself. Combined with AutoDock Vina, ∆-AEScore has an RMSE of 1.32 pK units and a Pearson’s correlation coefficient of 0.80 on the CASF-2016 benchmark, while retaining the good docking and screening power of the underlying classical scoring function. </p> </div> </div> </div>

MetalProGNet: a structure-based deep graph model for metalloprotein–ligand interaction predictions

“…42 The use of an AEV to predict molecular properties has been reported many times. 45–47 The calculation of the AEV is described in our previous publication. 42…”

“…42 The use of an AEV to predict molecular properties has been reported many times. [45][46][47] The calculation of the AEV is described in our previous publication. 42 All the above 3D geometrical-related features are rotationally and translationally invariant for the chemical system, where the rotation and translation invariance is oen absent in some interaction models represented by the 3D-CNN.…”

GNINA 1.0: Molecular Docking with Deep Learning

“…In order to define the side chains to be treated as flexible, autobox ligand is also used as flexdist ligand and flexdist is set to 3.5 Å, which gives a reasonable representation of the protein-ligand binding site. 57 Therefore, conformations for all side chains with at least one atom within 3.5 Å from flexdist ligand are sampled during docking.…”