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:

Predicting adverse drug reactions for combination pharmacotherapy with cross-scale associative learning via attention modules

Nature Computational Science volume 5, pages 547–561 (2025)Cite this article

Subjects

This article has been updated

Abstract

The rapid emergence of combination pharmacotherapies offers substantial therapeutic advantages but also poses risks of adverse drug reactions (ADRs). The accurate prediction of ADRs with interpretable computational methods is crucial for clinical medication management, drug development and precision medicine. Machine-learning and recently developed deep learning architectures struggle to effectively elucidate the key protein–protein interactions underlying ADRs from an organ perspective and to explicitly represent ADR associations. Here we propose OrganADR, an associative learning-enhanced model to predict ADRs at the organ level for emerging combination pharmacotherapy. It incorporates ADR information at the organ level, drug information at the molecular level and network-based biomedical knowledge into integrated representations with multi-interpretable modules. Evaluation across 15 organs demonstrates that OrganADR not only achieves state-of-the-art performance but also delivers both interpretable insights at the organ level and network-based perspectives. Overall, OrganADR represents a useful tool for cross-scale biomedical information integration and could be used to prevent ADRs during clinical precision medicine.

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

Access options

Buy this article

  • Purchase on SpringerLink
  • Instant access to the full article PDF.

USD 39.95

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

Fig. 1: Problem formulation: ADR prediction at organ level.
Fig. 2: Architecture of OrganADR.
Fig. 3: Construction of ADR-related datasets at the organ level.
Fig. 4: Information at the organ level facilitates ADR prediction.
Fig. 5: Identification of key PPIs using OrganADR.
Fig. 6: Comparison of OrganADR with baseline models.

Similar content being viewed by others

Data availability

This study utilizes TWOSIDES5 (https://tatonettilab.org/resources/tatonetti-stm.html), DrugBank26 (https://go.drugbank.com/releases/latest), Hetionet24 (https://github.com/hetio/hetionet), PrimeKG25 (https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/IXA7BM) and ADReCS6 (https://bioinf.xmu.edu.cn/ADReCS/download.jsp) as raw data. The preprocessed data for training, validation and testing OrganADR and baseline models are available via Zenodo at https://zenodo.org/records/15129153 (ref. 48). Source data are available with this paper.

Code availability

OrganADR (v.1.0.0) was developed to conduct data analysis, which is available via GitHub at https://github.com/BoyangLi-BIT/OrganADR (ref. 49) and Zenodo at https://zenodo.org/records/15129131 (ref. 48). Two-tailed hypothesis testing was performed using scipy.stats package (v.1.11.4) in Python (v.3.10.14).

Change history

  • 30 July 2025

    In the version of the article initially published, the Acknowledgements section omitted a funding resource, National Natural Science Foundation of China (grant no. 82202072). This is now amended in the HTML and PDF versions of the article.

References

  1. Leinonen, H. et al. A combination treatment based on drug repurposing demonstrates mutation-agnostic efficacy in pre-clinical retinopathy models. Nat. Commun. 15, 5943 (2024).

    Article  Google Scholar 

  2. Khunsriraksakul, C. et al. Integrating 3D genomic and epigenomic data to enhance target gene discovery and drug repurposing in transcriptome-wide association studies. Nat. Commun. 13, 3258 (2022).

    Article  Google Scholar 

  3. Lim, G. B. Benefits of combination pharmacotherapy for HFrEF. Nat. Rev. Cardiol. 17, 455 (2020).

    Article  Google Scholar 

  4. Jaaks, P. et al. Effective drug combinations in breast, colon and pancreatic cancer cells. Nature 603, 166–173 (2022).

    Article  Google Scholar 

  5. Tatonetti, N. P., Ye, P. P., Daneshjou, R. & Altman, R. B. Data-driven prediction of drug effects and interactions. Sci. Transl. Med. 4, 125ra31 (2012).

    Article  Google Scholar 

  6. Yue, Q.-X. et al. Mining real-world big data to characterize adverse drug reaction quantitatively: mixed methods study. J. Med. Internet Res. 26, 48572 (2024).

    Article  Google Scholar 

  7. Bouvy, J. C., De Bruin, M. L. & Koopmanschap, M. A. Epidemiology of adverse drug reactions in europe: a review of recent observational studies. Drug Safety 38, 437–453 (2015).

    Article  Google Scholar 

  8. Boland, M. R. et al. Systems biology approaches for identifying adverse drug reactions and elucidating their underlying biological mechanisms. Wiley Interdiscip. Rev. Syst. Biol. Med. 8, 104–122 (2016).

    Article  Google Scholar 

  9. Ivanov, S. M., Lagunin, A. A. & Poroikov, V. V. In silico assessment of adverse drug reactions and associated mechanisms. Drug Discov. Today 21, 58–71 (2016).

    Article  Google Scholar 

  10. Skou, S. T. et al. Multimorbidity. Nat. Rev. Dis. Primers 8, 48 (2022).

    Article  Google Scholar 

  11. Dewulf, P., Stock, M. & De Baets, B. Cold-start problems in data-driven prediction of drug–drug interaction effects. Pharmaceuticals 14, 429 (2021).

    Article  Google Scholar 

  12. Lee, C. Y. & Chen, Y.-P. P. Machine learning on adverse drug reactions for pharmacovigilance. Drug Discov. Today 24, 1332–1343 (2019).

    Article  Google Scholar 

  13. Lee, C. Y. & Chen, Y.-P. P. Prediction of drug adverse events using deep learning in pharmaceutical discovery. Brief. Bioinform. 22, 1884–1901 (2021).

    Article  Google Scholar 

  14. Rezaei, Z., Ebrahimpour-Komleh, H., Eslami, B., Chavoshinejad, R. & Totonchi, M. Adverse drug reaction detection in social media by deep learning methods. Cell J. 22, 319 (2020).

    Google Scholar 

  15. Zhao, H. et al. Identifying the serious clinical outcomes of adverse reactions to drugs by a multi-task deep learning framework. Commun. Biol. 6, 870 (2023).

    Article  Google Scholar 

  16. Zitnik, M., Agrawal, M. & Leskovec, J. Modeling polypharmacy side effects with graph convolutional networks. Bioinformatics 34, 457–466 (2018).

    Article  Google Scholar 

  17. Karim, M. R. et al. Drug–drug interaction prediction based on knowledge graph embeddings and convolutional-lstm network. In Proc. 10th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics (eds Shi, X. & Buck, M.) 113–123 (Association for Computing Machinery, 2019).

  18. Yu, Y. et al. Sumgnn: multi-typed drug interaction prediction via efficient knowledge graph summarization. Bioinformatics 37, 2988–2995 (2021).

    Article  Google Scholar 

  19. Zhang, Y. et al. Emerging drug interaction prediction enabled by a flow-based graph neural network with biomedical network. Nat. Comput. Sci. 3, 1023–1033 (2023).

    Article  Google Scholar 

  20. Wan, G. et al. Multi-organ immune-related adverse events from immune checkpoint inhibitors and their downstream implications: a retrospective multicohort study. Lancet Oncol. 25, 1053–1069 (2024).

    Article  Google Scholar 

  21. Langenberg, C., Hingorani, A. D. & Whitty, C. J. Biological and functional multimorbidity-from mechanisms to management. Nat. Med. 29, 1649–1657 (2023).

    Article  Google Scholar 

  22. Lesort, T., George, T. & Rish, I. Continual learning in deep networks: an analysis of the last layer. Preprint at https://doi.org/10.48550/arXiv.2106.01834 (2021).

  23. Xi, X., Gao, F., Xu, J., Guo, F. & Jin, T. Modeling output-level task relatedness in multi-task learning with feedback mechanism. Preprint at https://doi.org/10.48550/arXiv.2404.00885 (2024).

  24. Himmelstein, D. S. et al. Systematic integration of biomedical knowledge prioritizes drugs for repurposing. eLife 6, 26726 (2017).

    Article  Google Scholar 

  25. Chandak, P., Huang, K. & Zitnik, M. Building a knowledge graph to enable precision medicine. Sci. Data 10, 67 (2023).

    Article  Google Scholar 

  26. Knox, C. et al. Drugbank 6.0: the drugbank knowledgebase for 2024. Nucleic Acids Res. 52, 1265–1275 (2024).

    Article  Google Scholar 

  27. Buning, J. W. et al. Pharmacokinetics of oral hydrocortisone—results and implications from a randomized controlled trial. Metabolism 71, 7–16 (2017).

    Article  Google Scholar 

  28. Nguewa, P. A. et al. Pentamidine is an antiparasitic and apoptotic drug that selectively modifies ubiquitin. Chem. Biodivers. 2, 1387–1400 (2005).

    Article  Google Scholar 

  29. Wood, G., Wetzig, N., Hogan, P. & Whitby, M. Survival from pentamidine induced pancreatitis and diabetes mellitus. Aust. N. Z. J. Med. 21, 341–342 (1991).

    Article  Google Scholar 

  30. Buchman, A. L. Side effects of corticosteroid therapy. J. Clin. Gastroenterol. 33, 289–294 (2001).

    Article  Google Scholar 

  31. Davies, N. M. & Anderson, K. E. Clinical pharmacokinetics of diclofenac: therapeutic insights and pitfalls. Clin. Pharmacokinet. 33, 184–213 (1997).

    Article  Google Scholar 

  32. Anstey, A. & Lear, J. T. Azathioprine: clinical pharmacology and current indications in autoimmune disorders. BioDrugs 9, 33–47 (1998).

    Article  Google Scholar 

  33. Mozayani, A. & Raymon, L. A. Handbook of Drug Interactions: A Clinical and Forensic Guide (Springer, 2011).

  34. Heymann, R. E. et al. A double-blind, randomized, controlled study of amitriptyline, nortriptyline and placebo in patients with fibromyalgia. an analysis of outcome measures. Clin. Exp. Rheumatol. 19, 697–702 (2001).

    Google Scholar 

  35. Mamoshina, P., Rodriguez, B. & Bueno-Orovio, A. Toward a broader view of mechanisms of drug cardiotoxicity. Cell Rep. Med. 2, 100216 (2021).

    Article  Google Scholar 

  36. Clewell, H. J. et al. Review and evaluation of the potential impact of age-and gender-specific pharmacokinetic differences on tissue dosimetry. Crit. Rev. Toxicol. 32, 329–389 (2002).

    Article  Google Scholar 

  37. Ye, L., Hou, C. & Liu, S. The role of metabolizing enzymes and transporters in antiretroviral therapy. Curr. Topics Med. Chem. 17, 340–360 (2017).

    Article  Google Scholar 

  38. Scarlett, Y. Medical management of fecal incontinence. Gastroenterology 126, 55–63 (2004).

    Article  Google Scholar 

  39. Ereshefsky, L. & Sloan, D. Drug–drug interactions with the use of psychotropic medications. CNS Spectr. 14, 1–8 (2009).

    Google Scholar 

  40. Tanvir, F., Islam, M. I. K. & Akbas, E. Predicting drug–drug interactions using meta-path based similarities. In Proc. 2021 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (eds Hallinan, J. et al.) 1–8 (IEEE, 2021).

  41. Jiang, H. et al. Adverse drug reactions and correlations with drug–drug interactions: a retrospective study of reports from 2011 to 2020. Front. Pharmacol. 13, 923939 (2022).

    Article  Google Scholar 

  42. Roitmann, E., Eriksson, R. & Brunak, S. Patient stratification and identification of adverse event correlations in the space of 1190 drug related adverse events. Front. Physiol. 5, 332 (2014).

    Article  Google Scholar 

  43. King, C. et al. Pharmacogenomic associations of adverse drug reactions in asthma: systematic review and research prioritisation. Pharmacogenomics J. 20, 621–628 (2020).

    Article  Google Scholar 

  44. Özçelik, R., Ruiter, S., Criscuolo, E. & Grisoni, F. Chemical language modeling with structured state space sequence models. Nat. Commun. 15, 6176 (2024).

    Article  Google Scholar 

  45. Zheng, S. et al. Pharmkg: a dedicated knowledge graph benchmark for bomedical data mining. Brief. Bioinform. 22, 344 (2021).

    Article  Google Scholar 

  46. Dong, J., Liu, J., Wei, Y., Huang, P. & Wu, Q. Megakg: toward an explainable knowledge graph for early drug development. Preprint at bioRxiv https://doi.org/10.1101/2024.03.27.586981 (2024).

  47. Brown, E. G., Wood, L. & Wood, S. The medical dictionary for regulatory activities (meddra). Drug Safety 20, 109–117 (1999).

    Article  Google Scholar 

  48. Li, B., Qi, Y., Li, B. & Li, X. OrganADR_Data_20250403. Zenodo https://doi.org/10.5281/zenodo.15129153 (2025).

  49. Li, B., Qi, Y., Li, B. & Li, X. OrganADR_Code_20250403. Zenodo https://doi.org/10.5281/zenodo.15129131 (2025).

Download references

Acknowledgements

This project is supported by National Natural Science Foundation of China (grant no. 82202072). This project is also supported by Beijing Institute of Technology Science and Technology Innovation Plan (grant no. LY2023-32) and Beijing Institute of Technology Science and Technology Innovation Plan, Science and Technology Support Special Program (grant no. 2024CX02046).

Author information

Authors and Affiliations

  1. School of Interdisciplinary Science, Beijing Institute of Technology, Beijing, China

    Boyang Li & Bo Li

  2. School of Medical Technology, Beijing Institute of Technology, Beijing, China

    Boyang Li & Xiaoqiong Li

  3. Beijing Key Laboratory for Separation and Analysis in Biomedicine and Pharmaceuticals, Beijing Institute of Technology, Beijing, China

    Boyang Li, Bo Li & Xiaoqiong Li

  4. College of Computer Science, Nankai University, Tianjin, China

    Yifan Qi

Authors
  1. Boyang Li
    View author publications

    Search author on:PubMed Google Scholar

  2. Yifan Qi
    View author publications

    Search author on:PubMed Google Scholar

  3. Bo Li
    View author publications

    Search author on:PubMed Google Scholar

  4. Xiaoqiong Li
    View author publications

    Search author on:PubMed Google Scholar

Contributions

Bo Li and X.L. conceived the idea and guided the research. Boyang Li developed the model. Bo Li and Boyang Li wrote the manuscript. Boyang Li and Y.Q. contributed to algorithm implementation and results analysis. Bo Li and X.L. revised the manuscript. All authors read and approved the paper.

Corresponding authors

Correspondence to Bo Li or Xiaoqiong Li.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Computational Science thanks the anonymous reviewers 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 Ablation study and performance with different ADRs’ association matrices and imbalanced datasets.

a, Ablation study of the ADRs’ association matrix and corresponding modules. Minima, maxima, mean and P-value are shown. Sample size is 100. Two-tailed paired t-test or Wilcoxon signed-rank test are conducted (determined by normality assessment via Shapiro-Wilk test) using scipy.stats package (version 1.11.4) in Python (version 3.10.14). b, Performance comparison under different ADRs’ association matrices. Sample size is 12. c, OrganADR’s ROC-AUC and accuracy under imbalanced datasets. For x = 0.25, 0.33, 0.67, 0.75, 1.33, 1.5, 3 and 4, sample size is 5. For x = 0.5 and 2, sample size is 10. For x = 1, sample size is 20. For b and c, central line (median), box boundaries (25th-75th percentiles), whiskers extending to minima/maxima within 1.5 interquartile range (IQR) are shown. Outliers are shown as hollow circles.

Source data

Supplementary information

Supplementary Information

Supplementary Sects. 1 and 2, Figs. 1–52 and Tables 1–35.

Reporting Summary

Source data

Source Data Fig. 3.

Statistical source data.

Source Data Fig. 4.

Statistical source data.

Source Data Fig. 5.

Statistical source data.

Source Data Fig. 6.

Statistical source data.

Source Data Extended Data Fig. 1.

Statistical source data.

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

Li, B., Qi, Y., Li, B. et al. Predicting adverse drug reactions for combination pharmacotherapy with cross-scale associative learning via attention modules. Nat Comput Sci 5, 547–561 (2025). https://doi.org/10.1038/s43588-025-00816-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • Version of record:

  • Issue date:

  • DOI: https://doi.org/10.1038/s43588-025-00816-7

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