Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
80 tokens/sec
GPT-4o
59 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
7 tokens/sec
GPT-4.1 Pro
50 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Learning to Describe for Predicting Zero-shot Drug-Drug Interactions (2403.08377v1)

Published 13 Mar 2024 in cs.CL

Abstract: Adverse drug-drug interactions~(DDIs) can compromise the effectiveness of concurrent drug administration, posing a significant challenge in healthcare. As the development of new drugs continues, the potential for unknown adverse effects resulting from DDIs becomes a growing concern. Traditional computational methods for DDI prediction may fail to capture interactions for new drugs due to the lack of knowledge. In this paper, we introduce a new problem setup as zero-shot DDI prediction that deals with the case of new drugs. Leveraging textual information from online databases like DrugBank and PubChem, we propose an innovative approach TextDDI with a LLM-based DDI predictor and a reinforcement learning~(RL)-based information selector, enabling the selection of concise and pertinent text for accurate DDI prediction on new drugs. Empirical results show the benefits of the proposed approach on several settings including zero-shot and few-shot DDI prediction, and the selected texts are semantically relevant. Our code and data are available at \url{https://github.com/zhufq00/DDIs-Prediction}.

PDF HTML Abstract
Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Bookmark Chat Bubble Oval Streamline Icon: https://streamlinehq.com Chat (Pro)
Definition Search Book Streamline Icon: https://streamlinehq.com
References (42)
  1. Will affective computing emerge from foundation models and general artificial intelligence? a first evaluation of chatgpt. IEEE Intelligent Systems, 38(2):15–23.
  2. Natural products in drug discovery: advances and opportunities. Nature reviews Drug discovery, 20(3):200–216.
  3. A multitask, multilingual, multimodal evaluation of chatgpt on reasoning, hallucination, and interactivity. Technical report, arXiv:2302.04023.
  4. Richard Bellman. 1957. A markovian decision process. Indiana Univ. Math. J., 6:679–684.
  5. Language models are few-shot learners. NeurIPS, 33:1877–1901.
  6. Jacob Cohen. 1960. A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20(1):37–46.
  7. Rlprompt: Optimizing discrete text prompts with reinforcement learning. In EMNLP, pages 3369–3391.
  8. I Ralph Edwards and Jeffrey K Aronson. 2000. Adverse drug reactions: definitions, diagnosis, and management. The lancet, 356(9237):1255–1259.
  9. Tom Fawcett. 2006. An introduction to roc analysis. Pattern recognition letters, 27(8):861–874.
  10. Exploring the feasibility of chatgpt for event extraction. Technical report, arXiv:2303.03836.
  11. A review of approaches for predicting drug–drug interactions based on machine learning. Frontiers in Pharmacology, 12:3966.
  12. Daniel S Himmelstein and Sergio E Baranzini. 2015. Heterogeneous network edge prediction: a data integration approach to prioritize disease-associated genes. PLoS computational biology, 11(7):e1004259.
  13. Is chatgpt a good translator? yes with gpt-4 as the engine. Technical report, arXiv:2301.08745.
  14. Drug-drug interaction prediction based on knowledge graph embeddings and convolutional-lstm network. In ACM-BCB, pages 113–123.
  15. Diederik P Kingma and Jimmy Ba. 2014. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980.
  16. Thomas N Kipf and Max Welling. 2016. Semi-supervised classification with graph convolutional networks. Technical report, arXiv:1609.02907.
  17. Gyoung Ho Lee and Kong Joo Lee. 2017. Automatic text summarization using reinforcement learning with embedding features. In IJCNLP (Volume 2: Short Papers), pages 193–197.
  18. The power of scale for parameter-efficient prompt tuning. In EMNLP, pages 3045–3059.
  19. KGNN: Knowledge graph neural network for drug-drug interaction prediction. In IJCAI, volume 380, pages 2739–2745.
  20. RoBERTa: A robustly optimized bert pretraining approach. Technical report, arXiv:1907.11692.
  21. Fantastically ordered prompts and where to find them: Overcoming few-shot prompt order sensitivity. In ACL (Volume 1: Long Papers), pages 8086–8098.
  22. P Margaret Savitha and M Pushpa Rani. 2022. A comprehensive survey of ai methods to predict adverse drug-drug interactions. Computational Vision and Bio-Inspired Computing: Proceedings of ICCVBIC 2021, pages 495–511.
  23. Bethany Percha and Russ B Altman. 2013. Informatics confronts drug-drug interactions. Trends in pharmacological sciences, 34(3):178–184.
  24. David Rogers and Mathew Hahn. 2010. Extended-connectivity fingerprints. Journal of Chemical Information and Modeling, 50(5):742–754.
  25. Takaya Saito and Marc Rehmsmeier. 2015. The precision-recall plot is more informative than the roc plot when evaluating binary classifiers on imbalanced datasets. PloS one, 10(3):e0118432.
  26. Proximal policy optimization algorithms. Technical report, arXiv:1707.06347.
  27. Evaluation of chatgpt as a question answering system for answering complex questions. Technical report, arXiv:2303.07992.
  28. Data-driven prediction of drug effects and interactions. Science Translational Medicine, 4(125).
  29. Composition-based multi-relational graph convolutional networks. In ICLR.
  30. Drugbank 5.0: a major update to the drugbank database for 2018. Nucleic Acids Research, 46(D1):D1074–D1082.
  31. The severity of adverse drug reactions and their influencing factors based on the adr monitoring center of henan province. Scientific Reports, 11(1):20402.
  32. Effective knowledge graph embeddings based on multidirectional semantics relations for polypharmacy side effects prediction. Bioinformatics, 38(8):2315–2322.
  33. Learning to selectively learn for weakly supervised paraphrase generation with model-based reinforcement learning. In NAACL, pages 1385–1395.
  34. Sumgnn: multi-typed drug interaction prediction via efficient knowledge graph summarization. Bioinformatics, 37(18):2988–2995.
  35. Tempera: Test-time prompting via reinforcement learning. arXiv preprint arXiv:2211.11890.
  36. Yongqi Zhang and Quanming Yao. 2022. Knowledge graph reasoning with relational digraph. In Proceedings of the ACM web conference 2022, pages 912–924.
  37. Bilinear scoring function search for knowledge graph learning. IEEE TPAMI, 45(2):1458–1473.
  38. Emerging drug interaction prediction enabled by flow-based graph neural network with biomedical network. Nature Computational Science.
  39. Calibrate before use: Improving few-shot performance of language models. In ICML, pages 12697–12706. PMLR.
  40. A generative approach for script event prediction via contrastive fine-tuning. Proceedings of the AAAI Conference on Artificial Intelligence, 37(11):14056–14064.
  41. A diffusion model for event skeleton generation. In Findings of the Association for Computational Linguistics: ACL 2023, pages 12630–12641, Toronto, Canada. Association for Computational Linguistics.
  42. Modeling polypharmacy side effects with graph convolutional networks. Bioinformatics, 34(13):i457–i466.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (5)
  1. Fangqi Zhu (6 papers)
  2. Yongqi Zhang (33 papers)
  3. Lei Chen (484 papers)
  4. Bing Qin (186 papers)
  5. Ruifeng Xu (66 papers)