The Next Era: Deep Learning in Pharmaceutical Research
Over the past decade we have witnessed the increasing sophistication of machine learning algorithms applied in daily use from internet searches, voice recognition, social network software to machine vision software in cameras, phones, robots and self-driving cars. Pharmaceutical research has also seen its fair share of machine learning developments. For example, applying such methods to mine the growing datasets that are created in drug discovery not only enables us to learn from the past but to predict a molecule’s properties and behavior in future. The latest machine learning algorithm garnering significant attention is deep learning, which is an artificial neural network with multiple hidden layers. Publications over the last 3 years suggest that this algorithm may have advantages over previous machine learning methods and offer a slight but discernable edge in predictive performance. The time has come for a balanced review of this technique but also to apply machine learning methods such as deep learning across a wider array of endpoints relevant to pharmaceutical research for which the datasets are growing such as physicochemical property prediction, formulation prediction, absorption, distribution, metabolism, excretion and toxicity (ADME/Tox), target prediction and skin permeation, etc. We also show that there are many potential applications of deep learning beyond cheminformatics. It will be important to perform prospective testing (which has been carried out rarely to date) in order to convince skeptics that there will be benefits from investing in this technique.
KEY WORDSartificial intelligence deep Learning drug discovery machine learning pharmaceutics
Absorption, distribution, metabolism, excretion/toxicology
Area under the curve
Drug induced liver injury
Human ether a-go-go related gene
Quantitative structure activity relationships
Support vector machines
- 2.Rost B, Radivojac P, Bromberg Y. Protein function in precision medicine: deep understanding with machine learning. FEBS Lett. 2016;590(15):2327–41.Google Scholar
- 16.Chow J-F. Things to try after useR! – Part 1: Deep Learning with H2O. 2016 Aug 8th Available from: http://www.r-bloggers.com/things-to-try-after-user-part-1-deep-learning-with-h2o/.
- 17.Anon. TensorFlow. 2016 Aug 8th. Available from: https://www.tensorflow.org/.
- 18.Anon. Deeplearning4j 2016 Aug 8th. Available from: http://deeplearning4j.org/.
- 19.Novet J. Facebook open-sources its cutting-edge deep learning tools. 2016 Aug 8th. Available from: http://venturebeat.com/2015/01/16/facebook-opens-up-about-more-of-its-cutting-edge-deep-learning-tools/.
- 20.Chintala S. FAIR open sources deep-learning modules for Torch. 2016 Aug 8th. Available from: https://research.facebook.com/blog/fair-open-sources-deep-learning-modules-for-torch/.
- 21.Linn A. Microsoft releases CNTK, its open source deep learning toolkit, on GitHub. 2016 Aug 8th. Available from: http://blogs.microsoft.com/next/2016/01/25/microsoft-releases-cntk-its-open-source-deep-learning-toolkit-on-github/#sm.00013j280xp1sdctrgg21w81es5ov.
- 23.Min S, Lee B, Yoon S. Deep learning in bioinformatics. Brief Bioinform. 2016. doi:10.1093/bib/bbw068.
- 30.Fakoor R, Ladhak F, Nazi A, Huber M. Using deep learning to enhance cancer diagnosis and classification. In: Proceeding of the 30th International conference on machine learning. Atlanta, GA: JMLR: W&CP; 2013.Google Scholar
- 41.Wang C, Liu J, Luo F, Tan Y. Pairwise input neural network for target-ligand interaction prediction. IEEE Int Conf Bioinf and Biomed. 2014:67-70. doi:10.1109/BIBM.2014.6999129.
- 56.Unterthiner T, Mayr A, Klambauer G, Hochreiter S. Toxicity prediction using deep learning. Available from: https://arxiv.org/pdf/1503.01445.pdf.
- 68.Vock DM, Wolfson J, Bandyopadhyay S, Adomavicius G, Johnson PE, Vazquez-Benitez G, et al. Adapting machine learning techniques to censored time-to-event health record data: A general-purpose approach using inverse probability of censoring weighting. J Biomed Inform. 2016;61:119–31.PubMedCrossRefGoogle Scholar
- 82.Anon. Special report: the return of the machinery question. In: The Economist; 2016 June 25th. Available from: http://www.economist.com/news/special-report/21700761-after-many-false-starts-artificialintelligence-has-taken-will-it-cause-mass.
- 92.Murnane K. What is deep learning and how is it useful? Forbes. Available from: http://www.forbes.com/sites/kevinmurnane/2016/04/01/what-is-deep-learning-and-how-is-it-useful/#715d1eaf10f0.
- 93.Murnane K. Thirteen companies that use deep learning to produce actionable results. Forbes. Available from: http://www.forbes.com/sites/kevinmurnane/2016/04/01/thirteen-companies-that-use-deep-learning-to-produce-actionable-results/#4e710eb07967.
- 95.Tetko IV, Novotarskyi S, Sushko I, Ivanov V, Petrenko AE, Dieden R, et al. Development of dimethyl sulfoxide solubility models using 163,000 molecules: using a domain applicability metric to select more reliable predictions. J Chem Inf Model. 2013;53(8):1990–2000.PubMedPubMedCentralCrossRefGoogle Scholar