Using Drug Expression Profiles and Machine Learning Approach for Drug Repurposing
The cost of new drug development has been increasing, and repurposing known medications for new indications serves as an important way to hasten drug discovery. One promising approach to drug repositioning is to take advantage of machine learning (ML) algorithms to learn patterns in biological data related to drugs and then link them up to the potential of treating specific diseases. Here we give an overview of the general principles and different types of ML algorithms, as well as common approaches to evaluating predictive performances, with reference to the application of ML algorithms to predict repurposing opportunities using drug expression data as features. We will highlight common issues and caveats when applying such models to repositioning. We also introduce resources of drug expression data and highlight recent studies employing such an approach to repositioning.
Key wordsDrug repositioning Machine learning Drug transcriptome Genomics Deep learning
This work is partially supported by the Lo Kwee-Seong Biomedical Research Fund and a Direct Grant from the Chinese University of Hong Kong to HCS.
- 5.Aliper A, Plis S, Artemov A, Ulloa A, Mamoshina P, Zhavoronkov A (2016) Deep learning applications for predicting pharmacological properties of drugs and drug repurposing using transcriptomic data. Mol Pharm 13(7):2524–2530. https://doi.org/10.1021/acs.molpharmaceut.6b00248 CrossRefPubMedPubMedCentralGoogle Scholar
- 6.Zhao K, So H-C (2018) Drug repositioning for schizophrenia and depression/anxiety disorders: A machine learning approach leveraging expression data. IEEE journal of biomedical and health informatics (in press)Google Scholar
- 7.Friedman J, Hastie T, Tibshirani R (2001) The elements of statistical learning, vol 1. Springer Series in Statistics, New YorkGoogle Scholar
- 11.Bishop CM (2006) Pattern recognition and machine learning. Springer, New YorkGoogle Scholar
- 12.Xie LW, He S, Wen YQ, Bo XC, Zhang ZN (2017) Discovery of novel therapeutic properties of drugs from transcriptional responses based on multi-label classification. Sci Rep 7. https://doi.org/10.1038/s41598-017-07705-8 ARTN 7136
- 16.Breiman, L. (1984). Classification and regression trees. Belmont, CA.: Wadsworth International GroupGoogle Scholar
- 24.Buitinck L, Louppe G, Blondel M, Pedregosa F, Mueller A, Grisel O, Grobler J (2013) API design for machine learning software: experiences from the scikit-learn project. arXiv preprint arXiv 1309:0238Google Scholar
- 25.Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Duchesnay E (2011) Scikit-learn: machine learning in python. J Mach Learn Res 12:2825–2830Google Scholar
- 27.Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15:1929–1958Google Scholar
- 35.Davis J, Mark G (2006) The relationship between Precision-Recall and ROC curves. In: Proceedings of the 23rd international conference on Machine learning. ACM, pp 233–240Google Scholar
- 40.Bergstra J, Bengio Y (2012) Random search for hyper-parameter optimization. J Mach Learn Res 13:281–305Google Scholar
- 41.Louppe G, Wehenkel L, Sutera A, Geurts P (2013) Understanding variable importances in forests of randomized trees. In: Advances in neural information processing systems, pp 431–439Google Scholar