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:

A novel machine learning framework for automated biomedical relation extraction from large-scale literature repositories

Abstract

Knowledge about the relations between biomedical entities (such as drugs and targets) is widely distributed in more than 30 million research articles and consistently plays an important role in the development of biomedical science. In this work, we propose a novel machine learning framework, named BERE, for automatically extracting biomedical relations from large-scale literature repositories. BERE uses a hybrid encoding network to better represent each sentence from both semantic and syntactic aspects, and employs a feature aggregation network to make predictions after considering all relevant statements. More importantly, BERE can also be trained without any human annotation via a distant supervision technique. Through extensive tests, BERE has demonstrated promising performance in extracting biomedical relations, and can also find meaningful relations that were not reported in existing databases, thus providing useful hints to guide wet-lab experiments and advance the biological knowledge discovery process.

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

Access options

Rent or buy this article

Prices vary by article type

from$1.95

to$39.95

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

Fig. 1: The architecture of BERE.
Fig. 2: Test results on the distantly supervised DTI dataset.
Fig. 3: The in vitro inhibitory activity of nintedanib against JAK2 and EGFR.

Similar content being viewed by others

Data availability

The DDI and DTI datasets used in this work can be found at https://github.com/haiya1994/BERE. The full dataset for discovering potential DTIs is available from the corresponding authors upon request.

Code availability

The source code of BERE can be downloaded from the GitHub repository at https://github.com/haiya1994/BERE or the Zenodo repository at https://doi.org/10.5281/zenodo.3757058. All other code may be obtained from the corresponding authors upon request.

References

  1. Wishart, D. S. et al. DrugBank: a comprehensive resource for in silico drug discovery and exploration. Nucleic Acids Res. 34, D668–D672 (2006).

    Article  Google Scholar 

  2. Mattingly, C. J., Colby, G. T., Forrest, J. N. & Boyer, J. L. The Comparative Toxicogenomics Database (CTD). Environ. Health Perspect. 111, 793–795 (2003).

    Article  Google Scholar 

  3. Kuhn, M., Letunic, I., Jensen, L. J. & Bork, P. The SIDER database of drugs and side effects. Nucleic Acids Res. 44, D1075–D1079 (2015).

    Article  Google Scholar 

  4. Oughtred, R. et al. BioGRID: a resource for studying biological interactions in yeast. Cold Spring Harbor Protoc. 2016, pdb.top080754 (2016).

  5. Wang, S. et al. Annotating gene sets by mining large literature collections with protein networks. In Proceedings of the Pacific Symposium on Biocomputing 601–613 (World Scientific, 2018).

  6. Wang, S. et al. Deep functional synthesis: a machine learning approach to gene functional enrichment. Preprint at https://doi.org/10.1101/824086 (2019).

  7. Magro, L., Moretti, U. & Leone, R. Epidemiology and characteristics of adverse drug reactions caused by drug–drug interactions. Expert Opin. Drug Saf. 11, 83–94 (2012).

    Article  Google Scholar 

  8. Yang, F., Xu, J. & Zeng, J. Drug–target interaction prediction by integrating chemical, genomic, functional and pharmacological data. In Proceedings of the Pacific Symposium on Biocomputing 2014 148–159 (World Scientific, 2014).

  9. Luo, Y. et al. A network integration approach for drug–target interaction prediction and computational drug repositioning from heterogeneous information. Nat. Commun. 8, 573 (2017).

    Article  Google Scholar 

  10. Wan, F., Hong, L., Xiao, A., Jiang, T. & Zeng, J. NeoDTI: neural integration of neighbor information from a heterogeneous network for discovering new drug–target interactions. Bioinformatics 35, 104–111 (2018).

    Article  Google Scholar 

  11. Percha, B. & Altman, R. B. A global network of biomedical relationships derived from text. Bioinformatics 34, 2614–2624 (2018).

    Article  Google Scholar 

  12. Verga, P., Strubell E. & McCallum, A. Simultaneously self-attending to all mentions for full-abstract biological relation extraction. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies Vol. 1, 872–884 (ACL, 2018).

  13. Zhang, Y. et al. A hybrid model based on neural networks for biomedical relation extraction. J. Biomed. Inform. 81, 83–92 (2018).

    Article  Google Scholar 

  14. Yu, K. et al. Automatic extraction of protein–protein interactions using grammatical relationship graph. BMC Med. Inform. Decis. Mak. 18, 42 (2018).

    Article  Google Scholar 

  15. Lim, S., Lee, K. & Kang, J. Drug drug interaction extraction from the literature using a recursive neural network. PLoS ONE 13, e0190926 (2018).

    Article  Google Scholar 

  16. Mintz, M., Bills, S., Snow, R. & Jurafsky, D. Distant supervision for relation extraction without labeled data. In Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of the AFNLP Vol. 2, 1003–1011 (ACL, 2009).

  17. Riedel, S., Yao, L. & McCallum, A. Modeling relations and their mentions without labeled text. In Proceedings of the Joint European Conference on Machine Learning and Knowledge Discovery in Databases 148–163 (Springer, 2010).

  18. Dietterich, T. G., Lathrop, R. H. & Lozano-Pérez, T. Solving the multiple instance problem with axis-parallel rectangles. Artif. Intell. 89, 31–71 (1997).

    Article  Google Scholar 

  19. Jat, S., Khandelwal, S. & Talukdar, P. Improving distantly supervised relation extraction using word and entity based attention. In Proceedings of the 6th Workshop on Automated Knowledge Base Construction (2017).

  20. Vashishth, S., Joshi, R., Prayaga, S. S., Bhattacharyya, C. & Talukdar, P. RESIDE: improving distantly-supervised neural relation extraction using side information. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing 1257–1266 (ACL, 2018).

  21. Zeng, D., Liu, K., Chen, Y. & Zhao, J. Distant supervision for relation extraction via piecewise convolutional neural networks. In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing 1753–1762 (ACL, 2015).

  22. Quirk, C. & Poon, H. Distant supervision for relation extraction beyond the sentence boundary. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics Vol. 1, 1171–1182 (ACL, 2017).

  23. Lin, Y., Shen, S., Liu, Z., Luan, H. & Sun, M. Neural relation extraction with selective attention over instances. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics Vol. 1, 2124–2133 (ACL, 2016).

  24. Zhou, P. et al. Attention-based bidirectional long short-term memory networks for relation classification. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics Vol. 2, 207–212 (ACL, 2016).

  25. Sun, X. et al. Drug–drug interaction extraction via recurrent hybrid convolutional neural networks with an improved focal loss. Entropy 21, 37 (2019).

    Article  Google Scholar 

  26. Socher, R. et al. Recursive deep models for semantic compositionality over a sentiment treebank. In Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing 1631–1642 (ACL, 2013).

  27. Iyyer, M., Boyd-Graber, J., Claudino, L., Socher, R. & DauméIII, H. A neural network for factoid question answering over paragraphs. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) 633–644 (ACL, 2014).

  28. Hashimoto, K., Miwa, M., Tsuruoka, Y. & Chikayama, T. Simple customization of recursive neural networks for semantic relation classification. In Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing 1372–1376 (ACL, 2013).

  29. Li, J., Luong, M. T., Jurafsky, D. & Hovy, E. When are tree structures necessary for deep learning of representations? In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing 2304–2314 (ACL, 2015).

  30. Bowman, S. R. et al. A fast unified model for parsing and sentence understanding. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics Vol. 1, 1466–1477 (ACL, 2016).

  31. Yogatama, D., Blunsom, P., Dyer, C., Grefenstette, E. & Ling, W. Learning to compose words into sentences with reinforcement learning. In Proceedings of the 5th Interational Conference on Learning Representations (2017).

  32. Maillard, J., Clark, S. & Yogatama, D. Jointly learning sentence embeddings and syntax with unsupervised Tree-LSTMs. Nat. Lang. Eng. 25, 433–449 (2019).

    Article  Google Scholar 

  33. Choi, J., Yoo, K. M. & Lee, S.-g. Learning to compose task-specific tree structures. In Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence 5094–5101 (AAAI, 2018).

  34. Wang, X., Girshick, R., Gupta, A. & He, K. Non-local neural networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 7794–7803 (IEEE, 2018).

  35. Vaswani, A. et al. Attention is all you need. In Proceedings of Advances in Neural Information Processing Systems 5998–6008 (NIPS, 2017).

  36. Zhao, Z., Yang, Z., Luo, L., Lin, H. & Wang, J. Drug drug interaction extraction from biomedical literature using syntax convolutional neural network. Bioinformatics 32, 3444–3453 (2016).

    Google Scholar 

  37. Liu, S., Tang, B., Chen, Q. & Wang, X. Drug-drug interaction extraction via convolutional neural networks. Comput. Math. Methods Med. 2016, 6918381 (2016).

    MATH  Google Scholar 

  38. Quan, C., Hua, L., Sun, X. & Bai, W. Multichannel convolutional neural network for biological relation extraction. Biomed Res. Int. 2016, 1850404 (2016).

    Google Scholar 

  39. Sahu, S. K. & Anand, A. Drug–drug interaction extraction from biomedical texts using long short-term memory network. J. Biomed. Inform. 86, 15–24 (2018).

    Article  Google Scholar 

  40. Zhou, D., Miao, L. & He, Y. Position-aware deep multi-task learning for drug–drug interaction extraction. Artif. Intell. Med. 87, 1–8 (2018).

    Article  Google Scholar 

  41. Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I. & Salakhutdinov, R. Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15, 1929–1958 (2014).

    MathSciNet  MATH  Google Scholar 

  42. Tolias, G., Sicre, R. & Jégou, H. Particular object retrieval with integral max-pooling of CNN activations. In Proceedings of the 4th International Conference on Learning Representations (2016).

  43. Liu, C. Y. et al. The tyrosine kinase inhibitor nintedanib activates SHP-1 and induces apoptosis in triple-negative breast cancer cells. Exp. Mol. Med. 49, e366 (2017).

    Article  Google Scholar 

  44. Kato, M. et al. Gastrointestinal adverse effects of nintedanib and the associated risk factors in patients with idiopathic pulmonary fibrosis. Sci. Rep. 9, 12062 (2019).

    Article  Google Scholar 

  45. XLFit 5.4.0.8 (IDBS, 2014); https://www.idbs.com/excelcurvefitting/xlfit-product/

  46. Herrero-Zazo, M., Segura-Bedmar, I., Martínez, P. & Declerck, T. The DDI corpus: an annotated corpus with pharmacological substances and drug–drug interactions. J. Biomed. Inform. 46, 914–920 (2013).

    Article  Google Scholar 

  47. Li, J. et al. BioCreative V CDR task corpus: a resource for chemical disease relation extraction. Database 2016, baw068 (2016).

  48. Krallinger, M. et al. Overview of the BioCreative VI chemical-protein interaction track. In Proceedings of the Sixth BioCreative Challenge Evaluation Workshop Vol. 1, 141–146 (2017).

  49. Honnibal, M. & Montani, I. spaCy 2.0.18 (2018); https://spacy.io/

  50. Pyysalo, S., Ginter, F., Moen, H., Salakoski, T. & Ananiadou, S. Word vectors (NLPLab, 2013); http://bio.nlplab.org/

  51. Pyysalo, S., Ginter, F., Moen, H., Salakoski, T. & Ananiadou, S. Distributional semantics resources for biomedical text processing. In Proceedings of the 5th International Symposium on Languages in Biology and Medicine 39–44 (2013).

  52. Mikolov, T., Chen, K., Corrado, G. & Dean, J. Efficient estimation of word representations in vector space. In Proceedings of the 1st International Conference on Learning Representations (2013).

  53. Tan, Z., Wang, M., Xie, J., Chen, Y. & Shi, X. Deep semantic role labeling with self-attention. In Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence 16725 (AAAI, 2018).

  54. He, K., Zhang, X., Ren, S. & Sun, J. J. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 770–778 (IEEE, 2016).

  55. Cho, K., Van Merriënboer, B., Bahdanau, D. & Bengio, Y. On the properties of neural machine translation: Encoder-decoder approaches. In Proceedings of SSST-8, Eighth Workshop on Syntax, Semantics and Structure in Statistical Translation 103–111 (ACL, 2014).

  56. Socher, R., Lin, C. C., Manning, C. & Ng, A. Y. Parsing natural scenes and natural language with recursive neural networks. In Proceedings of the 28th International Conference on Machine Learning (ICML-11) 129–136 (ACM, 2011).

  57. Tai, K. S., Socher, R. & Manning, C. D. Improved semantic representations from tree-structured long short-term memory networks. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing Vol. 1, 1556–1566 (ACL, 2015).

  58. Kokkinos, F. & Potamianos, A. Structural attention neural networks for improved sentiment analysis. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics Vol. 2, 586–591 (ACL, 2017).

  59. Jang, E., Gu, S. & Poole, B. Categorical reparameterization with gumbel-softmax. In Proceedings of the 5th International Conference on Learning Representations (2017).

  60. Nair, V. & Hinton, G. E. Rectified linear units improve restricted Boltzmann machines. In Proceedings of the 27th International Conference on Machine Learning (ICML-10) 807–814 (ACM, 2010).

  61. Kingma, D. P. & Ba, J. Adam: A method for stochastic optimization. In Proceedings of the 3rd International Conference on Learning Representations (2015).

Download references

Acknowledgements

We thank Z. Liu, T. Yang and H. Hu for their helpful discussions about this work. This work was supported in part by the National Natural Science Foundation of China (grants 61872216, 81630103 and 31900862), the Turing AI Institute of Nanjing and the Zhongguancun Haihua Institute for Frontier Information Technology.

Author information

Authors and Affiliations

Authors

Contributions

L.H., D.Z. and J.Z. conceived the concept. L.H. designed the methodology and performed experiments. L.H., J.L., S.L., T.J. and D.Z. analysed the results. H.Y. contributed to wet-lab experiments. L.H. and J.Z. wrote the paper. S.L., F.W., T.J. and D.Z. contributed to revision of the manuscript.

Corresponding authors

Correspondence to Dan Zhao or Jianyang Zeng.

Ethics declarations

Competing interests

J.Z. is founder and CTO of Silexon AI Technology Co. Ltd and has an equity interest.

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 Comparison of the precision-recall curve between BERE and its alternatives with other sentence aggregation strategies on the distantly supervised DTI dataset.

BERE+POOL and BERE+AVE adopt a max-pooling strategy and an average strategy to aggregate sentence representations, respectively. The legend on the top right contains area under precision-recall curve (AUPRC) and F1-score for each method.

Extended Data Fig. 2 The hyperparameter settings of BERE on different test datasets.

The learning rates were determined using a grid search among {0.0001, 0.0002, …, 0.001}. Other hyper-parameters were set empirically.

Extended Data Fig. 3 The basic statistics of the datasets used in our tests.

(a) The numbers of sentences annotated with five different types of DDI relations in the DDI’13 dataset. NA means no interaction. ADVICE means the recommended concomitant medication usage. EFFECT means that there exists a certain pharmacodynamic effect between two drugs. MECHANISM means that there exists a certain pharmacokinetic mechanism between two drugs. INT means that a DDI occurs without any additional information. (b) The numbers of bags of sentences annotated with six different types of DTI relations in the distantly supervised DTI dataset. NA means no interaction.Substrate means that the drug is what the target (that is, enzyme) acts upon. Inhibitor means that the drug binds to the target (that is, enzyme) and impede with the functioning of the target. Agonist/Antagonist means that the drug binds to the target (that is, receptor) and activates/blocks it to produce a biological response. Unknown means that there exists a certain relation between a drug–target pair, but the action mechanism is unknown in DrugBank. Other is a unified name of all the other types of interactions with fewer occurrences. The unlabelled set, which was mainly used for prediction, was collected from the PMC articles after excluding abstracts.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Hong, L., Lin, J., Li, S. et al. A novel machine learning framework for automated biomedical relation extraction from large-scale literature repositories. Nat Mach Intell 2, 347–355 (2020). https://doi.org/10.1038/s42256-020-0189-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s42256-020-0189-y

This article is cited by

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