Deep active learning with high structural discriminability for molecular mutagenicity prediction.
Journal
Communications biology
ISSN: 2399-3642
Titre abrégé: Commun Biol
Pays: England
ID NLM: 101719179
Informations de publication
Date de publication:
31 Aug 2024
31 Aug 2024
Historique:
received:
09
10
2023
accepted:
21
08
2024
medline:
1
9
2024
pubmed:
1
9
2024
entrez:
31
8
2024
Statut:
epublish
Résumé
The assessment of mutagenicity is essential in drug discovery, as it may lead to cancer and germ cells damage. Although in silico methods have been proposed for mutagenicity prediction, their performance is hindered by the scarcity of labeled molecules. However, experimental mutagenicity testing can be time-consuming and costly. One solution to reduce the annotation cost is active learning, where the algorithm actively selects the most valuable molecules from a vast chemical space and presents them to the oracle (e.g., a human expert) for annotation, thereby rapidly improving the model's predictive performance with a smaller annotation cost. In this paper, we propose muTOX-AL, a deep active learning framework, which can actively explore the chemical space and identify the most valuable molecules, resulting in competitive performance with a small number of labeled samples. The experimental results show that, compared to the random sampling strategy, muTOX-AL can reduce the number of training molecules by about 57%. Additionally, muTOX-AL exhibits outstanding molecular structural discriminability, allowing it to pick molecules with high structural similarity but opposite properties.
Identifiants
pubmed: 39217273
doi: 10.1038/s42003-024-06758-6
pii: 10.1038/s42003-024-06758-6
doi:
Substances chimiques
Mutagens
0
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
1071Subventions
Organisme : National Natural Science Foundation of China (National Science Foundation of China)
ID : 62103436
Organisme : Science and Technology Commission of Shanghai Municipality (Shanghai Municipal Science and Technology Commission)
ID : 22xtcx00300
Informations de copyright
© 2024. The Author(s).
Références
Ames, B. N., Lee, F. D. & Durston, W. E. Improved bacterial test system for detection and classification of mutagens and carcinogens. Proc. Natl Acad. Sci. USA 70, 782–786 (1973).
doi: 10.1073/pnas.70.3.782
pubmed: 4577135
pmcid: 433358
Mortelmans, K. & Zeiger, E. The Ames Salmonella/microsome mutagenicity assay. Mutat. Res. 455, 29–60 (2000).
doi: 10.1016/S0027-5107(00)00064-6
pubmed: 11113466
Kramer, J. A., Sagartz, J. E. & Morris, D. L. The application of discovery toxicology and pathology towards the design of safer pharmaceutical lead candidates. Nat. Rev. Drug Discov. 6, 636–649 (2007).
doi: 10.1038/nrd2378
pubmed: 17643090
Kumar, A., Dobrovolsky, V. N., Dhawan, A. & Shanker, R. Mutagenicity: Assays and Applications. (Academic Press, 2017).
Townsend, P. A. & Grayson, M. N. Density functional theory in the prediction of mutagenicity: a perspective. Chem. Res. Toxicol. 34, 179–188 (2020).
doi: 10.1021/acs.chemrestox.0c00113
pubmed: 32643924
pmcid: 7887799
Timperio, A. M., Kuiper, H. A. & Zolla, L. Identification of a furazolidone metabolite responsible for the inhibition of amino oxidases. Xenobiotica 33, 153–167 (2003).
doi: 10.1080/0049825021000038459
pubmed: 12623758
Ames, B. N., Bartsch, H., Miller, J. A. & Gurney, E. G. Carcinogens as frameshift mutagens - metabolites and derivatives of 2-acetylaminofluorene and other aromatic amine carcinogens. Proc. Natl Acad. Sci. USA 69, 3128–3132 (1972).
doi: 10.1073/pnas.69.11.3128
pubmed: 4564203
pmcid: 389719
Ames, B. N., Durston, W. E., Yamasaki, E. & Lee, F. D. Carcinogens are mutagens: a simple test system combining liver homogenates for activation and bacteria for detection. Proc. Natl Acad. Sci. USA 70, 2281–2285 (1973).
doi: 10.1073/pnas.70.8.2281
pubmed: 4151811
pmcid: 433718
McCann, J., Spingarn, N. E., Kobori, J. & Ames, B. N. Detection of carcinogens as mutagens: bacterial tester strains with R factor plasmids. Proc. Natl Acad. Sci. USA 72, 979–983 (1975).
doi: 10.1073/pnas.72.3.979
pubmed: 165497
pmcid: 432447
Galati, S. et al. VenomPred: a machine learning based platform for molecular toxicity predictions. Int. J. Mol. Sci. 23, 2105 (2022).
doi: 10.3390/ijms23042105
pubmed: 35216217
pmcid: 8877213
Hillebrecht, A. et al. Comparative evaluation of in silico systems for ames test mutagenicity prediction: scope and limitations. Chem. Res. Toxicol. 24, 843–854 (2011).
doi: 10.1021/tx2000398
pubmed: 21534561
Yang, X., Zhang, Z., Li, Q. & Cai, Y. Quantitative structure–activity relationship models for genotoxicity prediction based on combination evaluation strategies for toxicological alternative experiments. Sci. Rep. 11, 8030 (2021).
doi: 10.1038/s41598-021-87035-y
pubmed: 33850191
pmcid: 8044236
Hansen, K. et al. Benchmark data set for in silico prediction of Ames mutagenicity. J. Chem. Inf. Model. 49, 2077–2081 (2009).
doi: 10.1021/ci900161g
pubmed: 19702240
Chu, C. S. M., Simpson, J. D., O’Neill, P. M. & Berry, N. G. Machine learning-Predicting Ames mutagenicity of small molecules. J. Mol. Graph. Model. 109, 108011 (2021).
doi: 10.1016/j.jmgm.2021.108011
pubmed: 34555723
Fradkin, P. et al. A graph neural network approach for molecule carcinogenicity prediction. Bioinformatics 38, i84–i91 (2022).
doi: 10.1093/bioinformatics/btac266
pubmed: 35758812
pmcid: 9235510
Wu, Z. et al. Mining toxicity information from large amounts of toxicity data. J. Med. Chem. 64, 6924–6936 (2021).
doi: 10.1021/acs.jmedchem.1c00421
pubmed: 33961429
Martínez, M. J. et al. Multitask deep neural networks for ames mutagenicity prediction. J. Chem. Inf. Model. 62, 6342–6351 (2022).
doi: 10.1021/acs.jcim.2c00532
pubmed: 36066065
Shinada, N. K. et al. Optimizing machine-learning models for mutagenicity prediction through better feature selection. Mutagenesis 37, 191–202 (2022).
doi: 10.1093/mutage/geac010
pubmed: 35554560
Feeney, S. V., Lui, R., Guan, D. & Matthews, S. Multiple instance learning improves ames mutagenicity prediction for problematic molecular species. Chem. Res. Toxicol. 36, 1227–1237 (2023).
doi: 10.1021/acs.chemrestox.2c00372
pubmed: 37477941
Honma, M. et al. Improvement of quantitative structure-activity relationship (QSAR) tools for predicting Ames mutagenicity: outcomes of the Ames/QSAR International Challenge Project. Mutagenesis 34, 3–16 (2019).
doi: 10.1093/mutage/gey031
pubmed: 30357358
Li, S. et al. MutagenPred-GCNNs: a graph convolutional neural network-based classification model for mutagenicity prediction with data-driven molecular fingerprints. Interdiscip. Sci. 13, 25–33 (2021).
doi: 10.1007/s12539-020-00407-2
pubmed: 33506363
Gong, Y., Xue, D., Chuai, G., Yu, J. & Liu, Q. DeepReac+: deep active learning for quantitative modeling of organic chemical reactions. Chem. Sci. 12, 14459–14472 (2021).
doi: 10.1039/D1SC02087K
pubmed: 34880997
pmcid: 8580052
Settles, B. Active learning literature survey. Computer Sciences Technical Report, University of Wisconsin-Madison. http://digital.library.wisc.edu/1793/60660 (2009).
Gal, Y. & Ghahramani, Z. Dropout as a bayesian approximation: representing model uncertainty in deep learning. In Proc. 33rd International Conference on Machine Learning. 1050-1059 (PMLR, 2016).
Konyushkova, K., Sznitman, R. & Fua, P. Geometry in active learning for binary and multi-class image segmentation. Comput. Vis. Image Undstand. 182, 1–16 (2019).
doi: 10.1016/j.cviu.2019.01.007
Lv, X., Duan, F., Jiang, J.-J., Fu, X. & Gan, L. Deep active learning for surface defect detection. Sensors 20, 1650 (2020).
doi: 10.3390/s20061650
pubmed: 32188066
pmcid: 7146140
Eyke, N. S., Green, W. H. & Jensen, K. F. Iterative experimental design based on active machine learning reduces the experimental burden associated with reaction screening. React. Chem. Eng. 5, 1963–1972 (2020).
doi: 10.1039/D0RE00232A
Xie, W. et al. Integrated random negative sampling and uncertainty sampling in active learning improve clinical drug safety drug–drug interaction information retrieval. Front. Pharmacol. 11, 582470 (2021).
doi: 10.3389/fphar.2020.582470
pubmed: 34017245
pmcid: 8130007
Wang, H. et al. Scientific discovery in the age of artificial intelligence. Nature 620, 47–60 (2023).
doi: 10.1038/s41586-023-06221-2
pubmed: 37532811
Kulichenko, M. et al. Uncertainty-driven dynamics for active learning of interatomic potentials. Nat. Comput. Sci. 3, 230–239 (2023).
doi: 10.1038/s43588-023-00406-5
pubmed: 38177878
pmcid: 10766548
Cai, L., Wang, L., Fu, X. & Zeng, X. Active semisupervised model for improving the identification of anticancer peptides. ACS Omega 6, 23998–24008 (2021).
doi: 10.1021/acsomega.1c03132
pubmed: 34568678
pmcid: 8459422
Bressan, R. S., Camargo, G., Bugatti, P. H. & Saito, P. T. M. Exploring active learning based on representativeness and uncertainty for biomedical data classification. IEEE J. Biomed. Health Inform. 23, 2238–2244 (2019).
doi: 10.1109/JBHI.2018.2881155
pubmed: 30442623
Wu, L. et al. TOXRIC: a comprehensive database of toxicological data and benchmarks. Nucleic Acids Res. https://doi.org/10.1093/nar/gkac1074 (2022).
Probst, D. & Reymond, J. L. Visualization of very large high-dimensional data sets as minimum spanning trees. J. Cheminform. 12, 12 (2020).
doi: 10.1186/s13321-020-0416-x
pubmed: 33431043
pmcid: 7015965
Scheffer, T., Decomain, C. & Wrobel, S. Active hidden Markov models for information extraction. In Proc. International Symposium on Intelligent Data Analysis. 309–318 (Berlin, Heidelberg, 2001).
Li, X. & Guo, Y. Adaptive active learning for image classification. In Proc. 26th IEEE Conference on Computer Vision and Pattern Recognition. 859–866 (IEEE, 2021).
Huang, S., Wang, T., Xiong, H., Huan, J. & Dou, D. Semi-supervised active learning with temporal output discrepancy. In Proc. IEEE/CVF International Conference on Computer Vision. 3427–3436 (2021).
Sener, O. & Savarese, S. Active learning for convolutional neural networks: a Core-Set approach. In Proc. 6th International Conference on Learning Representations (2018).
Ding, X. et al. Active learning for drug design: a case study on the plasma exposure of orally administered drugs. J. Med. Chem. 64, 16838–16853 (2021).
doi: 10.1021/acs.jmedchem.1c01683
pubmed: 34779199
Evans, C. P., Fleshner, N., Fitzpatrick, J. M. & Zlotta, A. R. An evidence-based approach to understanding the pharmacological class effect in the management of prostatic diseases. BJU Int. 95, 743–749 (2005).
doi: 10.1111/j.1464-410X.2005.05390.x
pubmed: 15794775
Chacko, R. et al. Data based predictive models for odor perception. Sci. Rep. 10, 17136 (2020).
doi: 10.1038/s41598-020-73978-1
pubmed: 33051564
pmcid: 7553929
Tata, J. R. Thalidomide and induced amphibian metamorphosis. Nature 204, 939–940 (1964).
doi: 10.1038/204939a0
pubmed: 14248715
Willett, P., Barnard, J. M. & Downs, G. M. Chemical similarity searching. J. Chem. Inf. Comput. Sci. 38, 983–996 (1998).
doi: 10.1021/ci9800211
Reker, D. Practical considerations for active machine learning in drug discovery. Drug Discov. Today Technol. 32-33, 73–79 (2019).
doi: 10.1016/j.ddtec.2020.06.001
pubmed: 33386097
Xu, C. et al. In silico prediction of chemical Ames mutagenicity. J. Chem. Inf. Model. 52, 2840–2847 (2012).
doi: 10.1021/ci300400a
pubmed: 23030379
Tropsha, A. Best practices for QSAR model development, validation, and exploitation. Mol. Inform. 29, 476–488 (2010).
doi: 10.1002/minf.201000061
pubmed: 27463326
Heller, S. R., McNaught, A., Pletnev, I., Stein, S. & Tchekhovskoi, D. InChI, the IUPAC international chemical identifier. J. Cheminform. 7, 23 (2015).
doi: 10.1186/s13321-015-0068-4
pubmed: 26136848
pmcid: 4486400
Ren, P. et al. A survey of deep active learning. ACM Comput. Surv. 54, 1–40 (2021).
Yoo, D. & Kweon, I. S. Learning loss for active learning. In Proc. of the IEEE Conference on Computer Vision and Pattern Recognition. 93–102 (IEEE, 2019).