Deep learning large-scale drug discovery and repurposing | Nature Computational Science
Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Deep learning large-scale drug discovery and repurposing

Nature Computational Science volume 4, pages 600–614 (2024)Cite this article

Subjects

Abstract

Large-scale drug discovery and repurposing is challenging. Identifying the mechanism of action (MOA) is crucial, yet current approaches are costly and low-throughput. Here we present an approach for MOA identification by profiling changes in mitochondrial phenotypes. By temporally imaging mitochondrial morphology and membrane potential, we established a pipeline for monitoring time-resolved mitochondrial images, resulting in a dataset comprising 570,096 single-cell images of cells exposed to 1,068 United States Food and Drug Administration-approved drugs. A deep learning model named MitoReID, using a re-identification (ReID) framework and an Inflated 3D ResNet backbone, was developed. It achieved 76.32% Rank-1 and 65.92% mean average precision on the testing set and successfully identified the MOAs for six untrained drugs on the basis of mitochondrial phenotype. Furthermore, MitoReID identified cyclooxygenase-2 inhibition as the MOA of the natural compound epicatechin in tea, which was successfully validated in vitro. Our approach thus provides an automated and cost-effective alternative for target identification that could accelerate large-scale drug discovery and repurposing.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on SpringerLink
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Framework of deep learning-based MOA prediction by profiling temporal mitochondrial phenotype for large-scale drug discovery and repurposing.
Fig. 2: High-throughput acquisition of time-lapse images for temporal mitochondrial phenotypes.
Fig. 3: Drugs with varied MOAs exhibit diverse mitochondrial phenotypes.
Fig. 4: Deep learning approaches for temporal mitochondrial phenotypes recognition.
Fig. 5: Classification of 477 FDA-approved drugs into 38 MOAs on the basis of temporal mitochondrial phenotypes using MitoReID.
Fig. 6: MitoReID model for predicting FDA-approved drugs with known MOAs.

Similar content being viewed by others

Data availability

The dataset used to train the model and all the model weights are available via Zenodo64 at https://doi.org/10.5281/zenodo.12730131. Molecular structural data were obtained from the PDB database (http://www.rcsb.org/). Annotations of FDA-approved drugs were collected from Drugbank at https://go.drugbank.com/, ChEMBL at https://www.ebi.ac.uk/chembl/ and the Drug Repurposing Hub at https://www.broadinstitute.org/drug-repurposing-hub. Source Data are provided with this paper.

Code availability

Codes can be accessed via GitHub at https://github.com/liweim/MitoReID. A stable version of the code used in this work is available via Zenodo65 at https://doi.org/10.5281/zenodo.12726571.

References

  1. DiMasi, J. A., Grabowski, H. G. & Hansen, R. W. Innovation in the pharmaceutical industry: new estimates of R&D costs. J. Health Econ. 47, 20–33 (2016).

    Article  Google Scholar 

  2. Swinney, D. C. & Anthony, J. How were new medicines discovered? Nat. Rev. Drug Discov. 10, 507–519 (2011).

    Article  Google Scholar 

  3. Rask-Andersen, M., Almen, M. S. & Schioth, H. B. Trends in the exploitation of novel drug targets. Nat. Rev. Drug Discov. 10, 579–590 (2011).

    Article  Google Scholar 

  4. Lee, H. & Lee, J. W. Target identification for biologically active small molecules using chemical biology approaches. Arch. Pharm. Res. 39, 1193–1201 (2016).

    Article  Google Scholar 

  5. Ha, J. et al. Recent advances in identifying protein targets in drug discovery. Cell Chem. Biol. 28, 394–423 (2021).

    Article  Google Scholar 

  6. Boutros, M., Heigwer, F. & Laufer, C. Microscopy-based high-content screening. Cell 163, 1314–1325 (2015).

    Article  Google Scholar 

  7. Chandrasekaran, S. N. et al. Image-based profiling for drug discovery: due for a machine-learning upgrade? Nat. Rev. Drug Discov. 20, 145–159 (2021).

    Article  Google Scholar 

  8. Caicedo, J. C. et al. Data-analysis strategies for image-based cell profiling. Nat. Methods 14, 849–863 (2017).

    Article  Google Scholar 

  9. Way, G. P. et al. Morphology and gene expression profiling provide complementary information for mapping cell state. Cell Syst. 13, 911–923 e9 (2022).

    Article  Google Scholar 

  10. Funk, L. et al. The phenotypic landscape of essential human genes. Cell 185, 4634–4653 e22 (2022).

    Article  Google Scholar 

  11. Thyme, S. B. et al. Phenotypic landscape of schizophrenia-associated genes defines candidates and their shared functions. Cell 177, 478–491.e20 (2019).

    Article  Google Scholar 

  12. Simm, J. et al. Repurposing high-throughput image assays enables biological activity prediction for drug discovery. Cell Chem. Biol. 25, 611–618.e3 (2018).

    Article  Google Scholar 

  13. Nyffeler, J. et al. Bioactivity screening of environmental chemicals using imaging-based high-throughput phenotypic profiling. Toxicol. Appl. Pharmacol. 389, 114876 (2020).

    Article  Google Scholar 

  14. Pegoraro, G. & Misteli, T. High-throughput imaging for the discovery of cellular mechanisms of disease. Trends Genet. 33, 604–615 (2017).

    Article  Google Scholar 

  15. Bray, M. A. et al. Cell Painting, a high-content image-based assay for morphological profiling using multiplexed fluorescent dyes. Nat. Protoc. 11, 1757–1774 (2016).

    Article  Google Scholar 

  16. Hofmarcher, M. et al. Accurate prediction of biological assays with high-throughput microscopy images and convolutional networks. J. Chem. Inf. Model. 59, 1163–1171 (2019).

    Article  Google Scholar 

  17. Perlman, Z. E. et al. Multidimensional drug profiling by automated microscopy. Science 306, 1194–1198 (2004).

  18. Lin, J. R., Fallahi-Sichani, M. & Sorger, P. K. Highly multiplexed imaging of single cells using a high-throughput cyclic immunofluorescence method. Nat. Commun. 6, 8390 (2015).

    Article  Google Scholar 

  19. Nunnari, J. & Suomalainen, A. Mitochondria: in sickness and in health. Cell 148, 1145–1159 (2012).

    Article  Google Scholar 

  20. Russell, O. M. et al. Mitochondrial diseases: hope for the future. Cell 181, 168–188 (2020).

    Article  Google Scholar 

  21. Jangili, P. et al. DNA-damage-response-targeting mitochondria-activated multifunctional prodrug strategy for self-defensive tumor therapy. Angew. Chem. Int. Ed. 61, e202117075 (2022).

    Article  Google Scholar 

  22. Carelli, V. & Chan, D. C. Mitochondrial DNA: impacting central and peripheral nervous systems. Neuron 84, 1126–1142 (2014).

    Article  Google Scholar 

  23. Glancy, B. Visualizing mitochondrial form and function within the cell. Trends Mol. Med. 26, 58–70 (2020).

    Article  Google Scholar 

  24. Cretin, E. et al. High-throughput screening identifies suppressors of mitochondrial fragmentation in OPA1 fibroblasts. EMBO Mol. Med. 13, e13579 (2021).

    Article  Google Scholar 

  25. Varkuti, B. H. et al. Neuron-based high-content assay and screen for CNS active mitotherapeutics. Sci. Adv. 6, eaaw8702 (2020).

    Article  Google Scholar 

  26. Chandrasekharan, A. et al. A high-throughput real-time in vitro assay using mitochondrial targeted roGFP for screening of drugs targeting mitochondria. Redox Biol. 20, 379–389 (2019).

    Article  Google Scholar 

  27. Iannetti, E. F. et al. Multiplexed high-content analysis of mitochondrial morphofunction using live-cell microscopy. Nat. Protoc. 11, 1693–1710 (2016).

    Article  Google Scholar 

  28. Pereira, G. C. et al. Drug-induced cardiac mitochondrial toxicity and protection: from doxorubicin to carvedilol. Curr. Pharm. Des. 17, 2113–2129 (2011).

    Article  Google Scholar 

  29. Varga, Z. V. et al. Drug-induced mitochondrial dysfunction and cardiotoxicity. Am. J. Physiol. Heart. Circ. Physiol. 309, H1453–H1467 (2015).

    Article  Google Scholar 

  30. Stringer, C. et al. Cellpose: a generalist algorithm for cellular segmentation. Nat. Methods 18, 100–106 (2021).

    Article  Google Scholar 

  31. Cao, M. et al. Plant exosome nanovesicles (PENs): green delivery platforms. Mater. Horiz. 10, 3879–3894 (2023).

    Article  Google Scholar 

  32. Zhang, D. et al. Microalgae-based oral microcarriers for gut microbiota homeostasis and intestinal protection in cancer radiotherapy. Nat. Commun. 13, 1413 (2022).

    Article  Google Scholar 

  33. Ji, X. et al. Capturing functional two-dimensional nanosheets from sandwich-structure vermiculite for cancer theranostics. Nat. Commun. 12, 1124 (2021).

    Article  Google Scholar 

  34. Zhong, D. et al. Orally deliverable strategy based on microalgal biomass for intestinal disease treatment. Sci. Adv. 7, eabi9265 (2021).

    Article  Google Scholar 

  35. Chen, F. et al. The V-ATPases in cancer and cell death. Cancer Gene Ther. 29, 1529–1541 (2022).

    Article  Google Scholar 

  36. Rizzuto, R. et al. Mitochondria as sensors and regulators of calcium signalling. Nat. Rev. Mol. Cell Biol. 13, 566–578 (2012).

    Article  Google Scholar 

  37. Giorgi, C., Marchi, S. & Pinton, P. The machineries, regulation and cellular functions of mitochondrial calcium. Nat. Rev. Mol. Cell Biol. 19, 713–730 (2018).

    Article  Google Scholar 

  38. Schmitt, N., Grunnet, M. & Olesen, S. P. Cardiac potassium channel subtypes: new roles in repolarization and arrhythmia. Physiol. Rev. 94, 609–653 (2014).

    Article  Google Scholar 

  39. Lei, M. et al. Modernized classification of cardiac antiarrhythmic drugs. Circulation 138, 1879–1896 (2018).

    Article  Google Scholar 

  40. Zheng, Z., Zheng, L. & Yang, Y. A discriminatively learned CNN embedding for person reidentification. In ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM) Vol. 14, 1–20 (ACM, 2017).

  41. Luo, H. et al. Bag of tricks and a strong baseline for deep person re-identification. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (IEEE, 2019).

  42. Carreira, J. & A. Zisserman. Quo vadis, action recognition? A new model and the kinetics dataset. Proc. IEEE Conference on Computer Vision and Pattern Recognition 6299–6308 (IEEE, 2017).

  43. Hermans, A., Beyer, L. & Leibe, B. In defense of the triplet loss for person re-identification. Preprint at https://arxiv.org/abs/1703.07737 (2017).

  44. Wen, Y. et al. A discriminative feature learning approach for deep face recognition. In European Conference On Computer Vision 499–515 (Springer, 2016).

  45. Szegedy, C. et al. Rethinking the inception architecture for computer vision. Proc. IEEE Conference On Computer Vision And Pattern Recognition 2818–2826 (2016).

  46. Moon, H. & Phillips, P. J. Computational and performance aspects of PCA-based face-recognition algorithms. Perception 30, 303–321 (2001).

    Article  Google Scholar 

  47. Zheng, L. et al. Scalable person re-identification: a benchmark. In Proc. IEEE International Conference On Computer Vision 1116–1124 (IEEE, 2015).

  48. Atanasov, A. G. et al. Natural products in drug discovery: advances and opportunities. Nat. Rev. Drug Discov. 20, 200–216 (2021).

    Article  Google Scholar 

  49. Zhou, J. et al. Graph neural networks: a review of methods and applications. AI Open 1, 57–81 (2020).

    Article  Google Scholar 

  50. Santos, R. et al. A comprehensive map of molecular drug targets. Nat. Rev. Drug Discov. 16, 19–34 (2017).

    Article  Google Scholar 

  51. Corsello, S. M. et al. The drug repurposing hub: a next-generation drug library and information resource. Nat. Med. 23, 405–408 (2017).

    Article  Google Scholar 

  52. Zdrazil, B. et al. The ChEMBL Database in 2023: a drug discovery platform spanning multiple bioactivity data types and time periods. Nucleic Acids Res. 52, D1180–D1192 (2024).

    Article  Google Scholar 

  53. Wishart, D. S. et al. DrugBank 5.0: a major update to the DrugBank database for 2018. Nucleic Acids Res. 46, D1074–D1082 (2018).

    Article  Google Scholar 

  54. MetaXpress v.6.6 https://www.moleculardevices.com/products/cellular-imaging-systems/high-content-analysis/metaxpress (Molecular Devices, 2020).

  55. AutoDock v.4.2.6 https://autodock.scripps.edu/ (CCSB, 2014).

  56. ChemOffice v.19.0 https://revvitysignals.com/products/research/chemdraw (Revvity Signals, 2019).

  57. Goodsell, D. S. et al. RCSB Protein Data Bank: enabling biomedical research and drug discovery. Protein Sci. 29, 52–65 (2020).

    Article  Google Scholar 

  58. PyMOL v.2.5 https://pymol.org/ (Schrödinger, 2021).

  59. Zhang, S. et al. Discovery of herbacetin as a novel SGK1 inhibitor to alleviate myocardial hypertrophy. Adv. Sci. 9, e2101485 (2022).

    Article  Google Scholar 

  60. He, K. et al. Deep residual learning for image recognition. Proc. IEEE Conference On Computer Vision And Pattern Recognition 770–778 (IEEE, 2016).

  61. Schroff, F., Kalenichenko, D. & Philbin, P. FaceNet: a unified embedding for face recognition and clustering. In 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (IEEE, 2015).

  62. He, K. et al. Delving deep into rectifiers: surpassing human-level performance on imagenet classification. Proc. IEEE International Conference On Computer Vision 1026–1034 (IEEE, 2015).

  63. LeCun, Y. et al. Backpropagation applied to handwritten ZIP Code recognition. Neural Comput. 1, 541–551 (1989).

    Article  Google Scholar 

  64. Li, W., Yu, M. & Wang, Y. Data for Deep Learning Large-Scale Drug Discovery and Repurposing (Zenodo, 2024); https://doi.org/10.5281/zenodo.12730131

  65. Li, W. liweim/MitoReID: v1.0 (Zenodo, 2024); https://doi.org/10.5281/zenodo.12726571

Download references

Acknowledgements

We are grateful for the support from ZJU PII-Molecular Devices Joint Laboratory. We thank Zhejiang Lab for providing high-performance GPU servers for deep learning research. Images in the illustration were created using BioRender.com. Funding: National Key Research and Development Program of China (grant no. 2023YFC3502801 to Y.W.), National Natural Science Foundation of China (grant no. 82173941 to Y.W.), Fundamental Research Funds for Central Universities (grant no. 226-2024-00001 to Y.W.), ‘Pioneer’ and ‘Leading Goose’ R&D Program of Zhejiang (grant no. 2024C01020 to W.L.), Innovation Team and Talents Cultivation Program of National Administration of Traditional Chinese Medicine (grant no. ZYYCXTD-D-202002 to Y.W.).

Author information

Author notes

  1. These authors contributed equally: Min Yu, Weiming Li.

Authors and Affiliations

  1. College of Pharmaceutical Sciences, Zhejiang University, Hangzhou, China

    Min Yu, Yunru Yu, Yiyu Cheng & Yi Wang

  2. Zhejiang Lab, Hangzhou, China

    Weiming Li

  3. School of Basic Medical Sciences, Zhejiang Chinese Medical University, Hangzhou, China

    Yu Zhao

  4. John A. Paulson School of Engineering and Applied Sciences, Harvard University, Cambridge, MA, USA

    Lizhi Xiao

  5. Department of Physiology and Pharmacology, Karolinska Institute, Stockholm, Sweden

    Volker M. Lauschke

  6. State Key Laboratory of Chinese Medicine Modernization, Innovation Center of Yangtze River Delta, Zhejiang University, Jiaxing, China

    Yiyu Cheng & Yi Wang

  7. Department of Materials Science and Engineering, Stanford University, Stanford, CA, USA

    Xingcai Zhang

  8. Innovation Institute for Artificial Intelligence in Medicine of Zhejiang University, Hangzhou, China

    Yi Wang

  9. Center for system biology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA

    Yi Wang

Authors
  1. Min Yu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Weiming Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Yunru Yu
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Yu Zhao
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Lizhi Xiao
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Volker M. Lauschke
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Yiyu Cheng
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Xingcai Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Yi Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

X.Z., Y.W. and Y.C. conceived the study. Y.W., X.Z., M.Y., Y.Y. and Y.Z. designed the experimental scheme. M.Y. collected the data. W.L. performed image data processing, and the model training and prediction. M.Y. and W.L. wrote the original draft of the manuscript, whereas Y.W., X.Z., V.M.L., L.X. and Y.C reviewed and edited it.

Corresponding authors

Correspondence to Yiyu Cheng, Xingcai Zhang or Yi Wang.

Ethics declarations

Competing interests

The authors declare that they have no competing interests.

Peer review

Peer review information

Nature Computational Science thanks Paul Czodrowski, Shibiao Wan, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editor: Kaitlin McCardle, in collaboration with the Nature Computational Science team.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 Application of MitoReID in natural compounds to identify epicatechin as a cyclooxygenase-2 inhibitor.

(a) Flowchart illustrating the process for predicting MOAs of natural compounds from Traditional Chinese Medicine (TCM). (b) Predicted results for five natural compounds. The blue bars represent predicted outcomes that have been reported in other studies. Abbreviations: AChR, acetylcholine receptor; ACE, angiotensin converting enzyme; GluR, glucocorticoid receptor; SNRI, serotonin-norepinephrine reuptake inhibitor. (c) Molecular structure of epicatechin. (d) The inhibitory effects of epicatechin on COX-2. (e) A schematic diagram illustrating the binding between epicatechin and COX-2, generated through molecular docking. (f) The result of cellular thermal shift assay (CETSA). (g) The result of surface plasmon resonance (SPR) experiments. KD, dissociation constant; Ka, association rate constant; Kd, dissociation rate constant.

Source data

Supplementary information

Supplementary Information

Supplementary Notes 1–5, Figs. 1–6 and Tables 1–3.

Reporting Summary

Supplementary Data 1

Predicted results of eight novel drugs with known MOA.

Supplementary Data 2

Predicted MOAs of 60 natural compounds.

Supplementary Data 3

Drug annotation list.

Source data

Source Data Fig. 2

Unprocessed images for Fig. 2b,c,e,f, and statistical source data for Fig. 2d–f.

Source Data Fig. 3

Unprocessed images for Fig. 3a,c, and statistical source data for Fig. 3c,e,f,g.

Source Data Fig. 5

Statistical source data for Fig.5a,b.

Source Data Fig. 6

Unprocessed images and statistical Source Data for Fig. 6b.

Source Data Extended Data Fig. 1

Statistical source data for Extended Fig.1b,d,g, and unprocessed gels for Extended Fig. 1f.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Yu, M., Li, W., Yu, Y. et al. Deep learning large-scale drug discovery and repurposing. Nat Comput Sci 4, 600–614 (2024). https://doi.org/10.1038/s43588-024-00679-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s43588-024-00679-4

Search

Advanced search

Quick links

Nature Briefing: Translational Research

Sign up for the Nature Briefing: Translational Research newsletter — top stories in biotechnology, drug discovery and pharma.

Get what matters in translational research, free to your inbox weekly. Sign up for Nature Briefing: Translational Research