Abstract
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, for which it has not been explicitly trained. Therefore, we show that the model can be combined with the classical scoring function AutoDock Vina in the context of Delta-learning, where corrections to the AutoDock Vina scoring function are learned instead of the protein-ligand binding affinity itself. Combined with AutoDock Vina, Delta-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 docking and screening power of the underlying classical scoring function.
Highlights
Structure-based drug discovery exploits knowledge of protein structures to design novel and potent compounds for a specific target
We demonstrated that atomic environment vectors (AEVs) are a promising representation of the protein-ligand binding site amenable to machine learning-based predictions of the protein-ligand binding affinity, and of corrections to classical scoring functions
The results reported here for AEScore show similar or better performance than other state-of-the-art machine learning and deep learning methods on the comparative assessment of scoring functions (CASF)-2013 and CASF2016 benchmarks in binding affinity prediction
Summary
Structure-based drug discovery exploits knowledge of protein structures to design novel and potent compounds for a specific target. Many scoring functions belonging to the first three categories have been developed over the past decades [6,7,8,9,10] Despite their successes in reproducing the binding pose, a rapid and accurate prediction of the protein-ligand binding affinity remains a very challenging task [11]. Machine learning and deep learning scoring functions have consistently improved protein-ligand binding affinity predictions [12]. These improvements build on decades of quantitative structure-activity relationship (QSAR) modelling, where simpler representations and regressors were used [13, 14]. Deep learning architectures—which are outperforming standard algorithms in image recognition and natural language processing [15,16,17,18,19]—are under active research, as demonstrated by
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
