Version 1
: Received: 24 January 2022 / Approved: 25 January 2022 / Online: 25 January 2022 (07:53:01 CET)
Version 2
: Received: 16 February 2022 / Approved: 16 February 2022 / Online: 16 February 2022 (02:57:38 CET)
Version 3
: Received: 23 May 2022 / Approved: 23 May 2022 / Online: 23 May 2022 (11:16:49 CEST)
Fan, F. J.; Shi, Y. Effects of Data Quality and Quantity on Deep Learning for Protein-Ligand Binding Affinity Prediction. Bioorganic & Medicinal Chemistry, 2022, 72, 117003. https://doi.org/10.1016/j.bmc.2022.117003.
Fan, F. J.; Shi, Y. Effects of Data Quality and Quantity on Deep Learning for Protein-Ligand Binding Affinity Prediction. Bioorganic & Medicinal Chemistry, 2022, 72, 117003. https://doi.org/10.1016/j.bmc.2022.117003.
Fan, F. J.; Shi, Y. Effects of Data Quality and Quantity on Deep Learning for Protein-Ligand Binding Affinity Prediction. Bioorganic & Medicinal Chemistry, 2022, 72, 117003. https://doi.org/10.1016/j.bmc.2022.117003.
Fan, F. J.; Shi, Y. Effects of Data Quality and Quantity on Deep Learning for Protein-Ligand Binding Affinity Prediction. Bioorganic & Medicinal Chemistry, 2022, 72, 117003. https://doi.org/10.1016/j.bmc.2022.117003.
Abstract
Prediction of protein-ligand binding affinities is crucial for computational drug discovery. A number of deep learning approaches have been developed in recent years to improve the accuracy of such affinity prediction. While the predicting power of these systems have advanced to some degrees depending on the dataset used for model training and testing, the effects of the quality and quantity of the underlying data have not been thoroughly examined. In this study, we employed erroneous datasets and data subsets of different sizes, created from one of the largest databases of experimental binding affinities, to train and evaluate a deep learning system based on convolutional neural networks. Our results show that data quality and quantity do have significant impacts on the prediction performance of trained models. Depending on the variations in data quality and quantity, the performance discrepancies could be comparable to or even larger than those observed among different deep learning approaches. In particular, the presence of proteins during model training leads to a dramatic increase in prediction accuracy. This implies that continued accumulation of high-quality affinity data, especially for new protein targets, is indispensable for improving deep learning models to better predict protein-ligand binding affinities.
Keywords
binding affinity prediction; machine learning; data quality; data quantity; deep learning
Subject
Biology and Life Sciences, Biochemistry and Molecular Biology
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Received:
23 May 2022
Commenter:
Yun Shi
Commenter's Conflict of Interests:
Author
Comment:
Further revisions were made, including the effects of errors with normal distribution, the clarification of differences between this study and previous ones, and the reiteration of our findings that data could have larger impacts than algorithms for DL models to better predict protein-ligand binding affinities.
Commenter: Yun Shi
Commenter's Conflict of Interests: Author