The DOI matches its Crossref record; 22 of the 22 references it cites were found in a registry or an index.
Machine‐learning scoring functions for structure‐based virtual screening
Abstract
Molecular docking predicts whether and how small molecules bind to a macromolecular target using a suitable 3D structure. Scoring functions for structure‐based virtual screening primarily aim at discovering which molecules bind to the considered target when these form part of a library with a much higher proportion of non‐binders. Classical scoring functions are essentially models building a linear mapping between the features describing a protein–ligand complex and its binding label. Machine learning, a major subfield of artificial intelligence, can also be used to build fast supervised learning models for this task. In this review, we analyzed such machine‐learning scoring functions for structure‐based virtual screening in the period 2015–2019. We have discussed what the shortcomings of current benchmarks really mean and what valid alternatives have been employed. The latter retrospective studies observed that machine‐learning scoring functions were substantially more accurate, in terms of higher hit rates and potencies, than the classical scoring functions they were compared to. Several of these machine‐learning scoring functions were also employed in prospective studies, in which mid‐nanomolar binders with novel chemical structures were directly discovered without any potency optimization. We have thus highlighted the codes and webservers that are available to build or apply machine‐learning scoring functions to prospective structure‐based virtual screening studies. A discussion of prospects for future work completes this review. This article is categorized under:Computer and Information Science > Chemoinformatics
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
XGBoost: A Scalable Tree Boosting System
Chen, T., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939785 · Cited by 54,569
Trott, O., & Olson, A. J. (2010) · Journal of Computational Chemistry · doi:10.1002/jcc.21334 · Cited by 38,412
Friesner, R. A., Banks, J. L., Murphy, R. B., Halgren, T. A., Klicic, J. J., Mainz, D. T., Repasky, M. P., Knoll, E. H., Shelley, M., Perry, J. K., et al. (2004) · Journal of Medicinal Chemistry · doi:10.1021/jm0306430 · Cited by 10,057 · cited 2×
“82 by GScore.(57) The average values of EF1% and EF2% were increased from 27.”
BackgroundInfluentialSaito, T., & Rehmsmeier, M. (2015) · PLOS ONE · doi:10.1371/journal.pone.0118432 · Cited by 5,309
A few useful things to know about machine learning
Domingos, P. (2012) · Communications of the ACM · doi:10.1145/2347736.2347755 · Cited by 3,326 · cited 1×
“36,107 Conversely, it has been observed that larger training datasets result in a larger performance gap between ML-based SFs and classical SFs. 36 The limitations of established benchmarks, which affect any type of SF, are actually worse than it has been recognized.”
BackgroundMoleculeNet: a benchmark for molecular machine learning
Wu, Z., Ramsundar, B., Feinberg, E. N., Gomes, J., Geniesse, C., Pappu, A. S., Leswing, K., & Pande, V. (2018) · Chemical Science · doi:10.1039/c7sc02664a · Cited by 3,107
Directory of useful decoys, enhanced (DUD-E): better ligands and decoys for better benchmarking.
Mysinger, M. M., Carchia, M., Irwin, J. J., & Shoichet, B. K. (2012) · Journal of Medicinal Chemistry · doi:10.1021/jm300687e · Cited by 2,412 · cited 1×
“30 In reality, using property-matched decoys only reduces the SBVS performance of DOCK 3.6 from LogAUC 0.001 26.8% (randomly chosen decoys) to 24.4% (DUD-E decoys) over the 102 targets.”
MethodInfluentialBenchmarking sets for molecular docking.
Huang, N., Shoichet, B. K., & Irwin, J. J. (2006) · Journal of Medicinal Chemistry · doi:10.1021/jm0608356 · Cited by 1,318
Automated design of ligands to polypharmacological profiles
Besnard, J., Ruda, G. F., Setola, V., Abecassis, K., Rodriguiz, R. M., Huang, X.-P., Norval, S., Sassano, M. F., Shin, A. I., Webster, L. A., et al. (2012) · Nature · doi:10.1038/nature11691 · Cited by 832 · cited 1×
“The concept of multi-target drug development 71 is particularly attractive.”
BackgroundFrom machine learning to deep learning: progress in machine intelligence for rational drug discovery
Zhang, L., Tan, J., Han, D., & Zhu, H. (2017) · Drug Discovery Today · doi:10.1016/j.drudis.2017.08.010 · Cited by 760 · cited 1×
“Interesting reviews have also been published on the application of ML to quantitative structure – activity relationship (QSAR) modeling 42 as well as proteochemometric modeling.”
MethodLarge-scale comparison of machine learning methods for drug target prediction on ChEMBL
Mayr, A., Klambauer, G., Unterthiner, T., Steijaert, M., Wegner, J. K., Ceulemans, H., Clevert, D.-A., & Hochreiter, S. (2018) · Chemical Science · doi:10.1039/c8sc00148k · Cited by 584
Boosting docking-based virtual screening with deep learning.
Pereira, J. C., Caffarena, E. R., & dos Santos, C. N. (2016) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.6b00355 · Cited by 395
Performance of machine-learning scoring functions in structure-based virtual screening.
Wójcikowski, M., Ballester, P. J., & Siedlecki, P. (2017) · Scientific Reports · doi:10.1038/srep46710 · Cited by 364
Self-citationAin, Q. U., Aleksandrova, A., Roessler, F. D., & Ballester, P. J. (2015) · WIREs Computational Molecular Science · doi:10.1002/wcms.1225 · Cited by 357
Assessing the impact of generative AI on medicinal chemistry
Walters, W. P., & Murcko, M. (2020) · Nature Biotechnology · doi:10.1038/s41587-020-0418-2 · Cited by 262 · cited 2×
“126 There are several studies prior to the reviewed period, 46,54,123 where ML-based SFs also discovered new actives for other targets.”
BackgroundCang, Z., Mu, L., & Wei, G.-W. (2018) · PLOS Computational Biology · doi:10.1371/journal.pcbi.1005929 · Cited by 257
Li, H., Leung, K., Wong, M., & Ballester, P. J. (2015) · Molecular Informatics · doi:10.1002/minf.201400132 · Cited by 252
Self-citationLooking beyond the hype: applied AI and machine learning in translational medicine
Toh, T. S., Dondelinger, F., & Wang, D. (2019) · EBioMedicine · doi:10.1016/j.ebiom.2019.08.027 · Cited by 159
Baum, B., Muley, L., Smolinski, M., Heine, A., Hangauer, D., & Klebe, G. (2010) · Journal of Molecular Biology · doi:10.1016/j.jmb.2010.02.007 · Cited by 157
Beware of machine learning-based scoring functions-on the danger of developing black boxes.
Gabel, J., Desaphy, J., & Rognan, D. (2014) · Journal of Chemical Information and Modeling · doi:10.1021/ci500406k · Cited by 143 · cited 17×
“Section 3 overviews the development and retrospective validation of machine-learning SFs for SBVS. Section 4 concisely goes through freely available resources to develop and use ML-based SFs. Section 5 discusses the weaknesses of commonly used benchmarks for any SF, not just those employing ML. Section 6 reviews the p…”
BackgroundInfluentialEricksen, S. S., Wu, H., Zhang, H., Michael, L. A., Newton, M. A., Hoffmann, F. M., & Wildman, S. A. (2017) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.7b00153 · Cited by 107
Ding, B., Wang, J., Li, N., & Wang, W. (2013) · Journal of Chemical Information and Modeling · doi:10.1021/ci300508m · Cited by 45 · cited 3×
“Hou and co-workers presented a prospective application of MIEC-SVM, 52 a SF combining molecular interaction energy components (MIEC) derived from free energy decomposition and SVM. 53 A retrospective validation of MIEC-SVM on three kinase targets (ABL, ALK, and BRAF) was carried out first.”
MethodInfluential
Cited by
Artificial intelligence in drug discovery: recent advances and future perspectives
Jiménez-Luna, J., Grisoni, F., Weskamp, N., & Schneider, G. (2021) · Expert Opinion on Drug Discovery · doi:10.1080/17460441.2021.1909567 · Cited by 464
AI in drug discovery and its clinical relevance
Qureshi, R., Irfan, M., Gondal, T. M., Khan, S., Wu, J., Hadi, M. U., Heymach, J., Le, X., Yan, H., & Alam, T. (2023) · Heliyon · doi:10.1016/j.heliyon.2023.e17575 · Cited by 319
“In [22], the progress of traditional machine learning algorithms for protein-ligand docking scoring functions, and in [23] machine learning-based scoring function for structure-based virtual screening are presented.”
Cao, D., Chen, M., Zhang, R., Wang, Z., Huang, M., Yu, J., Jiang, X., Fan, Z., Zhang, W., Zhou, H., et al. (2025) · Nature Methods · doi:10.1038/s41592-024-02516-y · Cited by 106
Docking-Informed Machine Learning for Kinome-wide Affinity Prediction
Schifferstein, J., Bernatavicius, A., & Janssen, A. P. A. (2024) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.4c01260 · Cited by 8
Using Stochastic Computing for Virtual Screening Acceleration
Frasser, C. F., de Benito, C., Skibinsky-Gitlin, E. S., Canals, V., Font-Rosselló, J., Roca, M., Ballester, P. J., & Rosselló, J. L. (2021) · Electronics · doi:10.3390/electronics10232981 · Cited by 3
Self-citation
Metadata
- Authors
- Li, H., Sze, K., Lu, G., & Ballester, P. J.
- Venue
- WIREs Computational Molecular Science
- Publisher
- Wiley
- Published
- 2020-04-22
- Volume
- 11
- Issue
- 1
- ISSN
- 1759-0876 (ISSN-L)
- 1759-0884
- OpenAlex
- W3020257619
- Cited by
- 163
- References
- 22
- FWCI
- 12.03
- Type
- journal article
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.