Machine learning and ligand binding predictions: A review of data, methods, and obstacles
- PMID: 32057823
- DOI: 10.1016/j.bbagen.2020.129545
Machine learning and ligand binding predictions: A review of data, methods, and obstacles
Abstract
Computational predictions of ligand binding is a difficult problem, with more accurate methods being extremely computationally expensive. The use of machine learning for drug binding predictions could possibly leverage the use of biomedical big data in exchange for time-intensive simulations. This paper reviews current trends in the use of machine learning for drug binding predictions, data sources to develop machine learning algorithms, and potential problems that may lead to overfitting and ungeneralizable models. A few popular datasets that can be used to develop virtual high-throughput screening models are characterized using spatial statistics to quantify potential biases. We can see from evaluating some common benchmarks that good performance correlates with models with high-predicted bias scores and models with low bias scores do not have much predictive power. A better understanding of the limits of available data sources and how to fix them will lead to more generalizable models that will lead to novel drug discovery.
Keywords: Drug binding; Drug discovery; Machine learning; Overfitting.
Copyright © 2020 Elsevier B.V. All rights reserved.
Similar articles
-
Machine learning in computational docking.Artif Intell Med. 2015 Mar;63(3):135-52. doi: 10.1016/j.artmed.2015.02.002. Epub 2015 Feb 16. Artif Intell Med. 2015. PMID: 25724101
-
Novel Big Data-Driven Machine Learning Models for Drug Discovery Application.Molecules. 2022 Jan 18;27(3):594. doi: 10.3390/molecules27030594. Molecules. 2022. PMID: 35163865 Free PMC article.
-
Artificial Intelligence, Big Data and Machine Learning Approaches in Precision Medicine & Drug Discovery.Curr Drug Targets. 2021;22(6):631-655. doi: 10.2174/1389450122999210104205732. Curr Drug Targets. 2021. PMID: 33397265 Review.
-
Machine Learning Methods in Drug Discovery.Molecules. 2020 Nov 12;25(22):5277. doi: 10.3390/molecules25225277. Molecules. 2020. PMID: 33198233 Free PMC article. Review.
-
Most Ligand-Based Classification Benchmarks Reward Memorization Rather than Generalization.J Chem Inf Model. 2018 May 29;58(5):916-932. doi: 10.1021/acs.jcim.7b00403. Epub 2018 May 8. J Chem Inf Model. 2018. PMID: 29698607
Cited by
-
SMPLIP-Score: predicting ligand binding affinity from simple and interpretable on-the-fly interaction fingerprint pattern descriptors.J Cheminform. 2021 Mar 25;13(1):28. doi: 10.1186/s13321-021-00507-1. J Cheminform. 2021. PMID: 33766140 Free PMC article.
-
Green chemistry and coronavirus.Sustain Chem Pharm. 2021 Jun;21:100415. doi: 10.1016/j.scp.2021.100415. Epub 2021 Mar 3. Sustain Chem Pharm. 2021. PMID: 33686371 Free PMC article. Review.
-
Molecular dynamics: a powerful tool for studying the medicinal chemistry of ion channel modulators.RSC Med Chem. 2021 Jul 22;12(9):1503-1518. doi: 10.1039/d1md00140j. eCollection 2021 Sep 23. RSC Med Chem. 2021. PMID: 34671734 Free PMC article. Review.
-
Artificial intelligence in the prediction of protein-ligand interactions: recent advances and future directions.Brief Bioinform. 2022 Jan 17;23(1):bbab476. doi: 10.1093/bib/bbab476. Brief Bioinform. 2022. PMID: 34849575 Free PMC article. Review.
-
OnionNet-2: A Convolutional Neural Network Model for Predicting Protein-Ligand Binding Affinity Based on Residue-Atom Contacting Shells.Front Chem. 2021 Oct 27;9:753002. doi: 10.3389/fchem.2021.753002. eCollection 2021. Front Chem. 2021. PMID: 34778208 Free PMC article.
Publication types
MeSH terms
Substances
Grants and funding
LinkOut - more resources
Full Text Sources
