Discovery of novel therapeutic properties of drugs from transcriptional responses based on multi-label classification

Drug repositioning strategies have improved substantially in recent years. At present, two advances are poised to facilitate new strategies. First, the LINCS project can provide rich transcriptome data that reflect the responses of cells upon exposure to various drugs. Second, machine learning algorithms have been applied successfully in biomedical research. In this paper, we developed a systematic method to discover novel indications for existing drugs by approaching drug repositioning as a multi-label classification task and used a Softmax regression model to predict previously unrecognized therapeutic properties of drugs based on LINCS transcriptome data. This approach to complete the said task has not been achieved in previous studies. By performing in silico comparison, we demonstrated that the proposed Softmax method showed markedly superior performance over those of other methods. Once fully trained, the method showed a training accuracy exceeding 80% and a validation accuracy of approximately 70%. We generated a highly credible set of 98 drugs with high potential to be repositioned for novel therapeutic purposes. Our case studies included zonisamide and brinzolamide, which were originally developed to treat indications of the nervous system and sensory organs, respectively. Both drugs were repurposed to the cardiovascular category.

[1]  General pharmacology of the novel antiepileptic compound zonisamide. 2nd communication: effects on cardiovascular, visceral, renal and blood functions. , 1987, Arzneimittel-Forschung.

[2]  Bruce L. Booth,et al.  Quest for the best , 2003, Nature Reviews Drug Discovery.

[3]  Y. Sasaki,et al.  Faster evolution and evolvability control of genetic algorithms using a Softmax Mutation method , 2003, The 2003 Congress on Evolutionary Computation, 2003. CEC '03..

[4]  R. W. Hansen,et al.  The price of innovation: new estimates of drug development costs. , 2003, Journal of health economics.

[5]  R. Frank,et al.  New estimates of drug development costs. , 2003, Journal of health economics.

[6]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[7]  T. Ashburn,et al.  Drug repositioning: identifying and developing new uses for existing drugs , 2004, Nature Reviews Drug Discovery.

[8]  P. Brugada,et al.  Brugada syndrome: From cell to bedside , 2001 .

[9]  Chih-Jen Lin,et al.  A tutorial on?-support vector machines , 2005 .

[10]  Van V. Brantner,et al.  Estimating the cost of new drug development: is it really 802 million dollars? , 2006, Health affairs.

[11]  Paul A Clemons,et al.  The Connectivity Map: Using Gene-Expression Signatures to Connect Small Molecules, Genes, and Disease , 2006, Science.

[12]  A. Grace,et al.  Scn3b knockout mice exhibit abnormal ventricular electrophysiological properties , 2008, Progress in biophysics and molecular biology.

[13]  Dirk Van den Poel,et al.  FACULTEIT ECONOMIE , 2007 .

[14]  M. Boguski,et al.  Repurposing with a Difference , 2009, Science.

[15]  Diego di Bernardo,et al.  Identifying Network of Drug Mode of Action by Gene Expression Profiling , 2009, J. Comput. Biol..

[16]  Michael J. Keiser,et al.  Predicting new molecular targets for known drugs , 2009, Nature.

[17]  R. Tagliaferri,et al.  Discovery of drug mode of action and drug repositioning from transcriptional responses , 2010, Proceedings of the National Academy of Sciences.

[18]  P. Aloy,et al.  Unveiling the role of network and systems biology in drug discovery. , 2010, Trends in pharmacological sciences.

[19]  Cheng Zhu,et al.  Drug repositioning for orphan diseases , 2011, Briefings Bioinform..

[20]  Damian Szklarczyk,et al.  STITCH 3: zooming in on protein–chemical interactions , 2011, Nucleic Acids Res..

[21]  Jia Liu,et al.  Candesartan acutely recruits skeletal and cardiac muscle microvasculature in healthy humans. , 2012, The Journal of clinical endocrinology and metabolism.

[22]  Yan Zhao,et al.  Drug repositioning: a machine-learning approach through data integration , 2013, Journal of Cheminformatics.

[23]  David S. Wishart,et al.  DrugBank 4.0: shedding new light on drug metabolism , 2013, Nucleic Acids Res..

[24]  Min-Ling Zhang,et al.  A Review on Multi-Label Learning Algorithms , 2014, IEEE Transactions on Knowledge and Data Engineering.

[25]  C. Indolfi,et al.  Carbonic Anhydrase Activation Is Associated With Worsened Pathological Remodeling in Human Ischemic Diabetic Cardiomyopathy , 2014, Journal of the American Heart Association.

[26]  Geoffrey E. Hinton,et al.  Distilling the Knowledge in a Neural Network , 2015, ArXiv.

[27]  Leyi Wei,et al.  An Improved Protein Structural Classes Prediction Method by Incorporating Both Sequence and Structure Information. , 2015, IEEE transactions on nanobioscience.

[28]  William Stafford Noble,et al.  Machine learning applications in genetics and genomics , 2015, Nature Reviews Genetics.

[29]  Xing Gao,et al.  Enhanced Protein Fold Prediction Method Through a Novel Feature Extraction Technique , 2015, IEEE Transactions on NanoBioscience.

[30]  Ralph Mazitschek,et al.  Treatment of Obesity with Celastrol , 2015, Cell.

[31]  Masaru Sugimachi,et al.  From Cell to Bedside , 2015 .

[32]  Q. Zou,et al.  A novel machine learning method for cytokine-receptor interaction prediction. , 2016, Combinatorial chemistry & high throughput screening.

[33]  Gang Fu,et al.  PubChem Substance and Compound databases , 2015, Nucleic Acids Res..

[34]  Q. Zou,et al.  Recent Progress in Machine Learning-Based Methods for Protein Fold Recognition , 2016, International journal of molecular sciences.

[35]  Peer Bork,et al.  The SIDER database of drugs and side effects , 2015, Nucleic Acids Res..

[36]  P Vallotton,et al.  Detection of tubule boundaries based on circular shortest path and polar‐transformation of arbitrary shapes , 2016, Journal of microscopy.

[37]  Zhiyong Chen,et al.  Exploring local discriminative information from evolutionary profiles for cytokine-receptor interaction prediction , 2016, Neurocomputing.

[38]  Zhongkui Hong,et al.  Mechanical activation of angiotensin II type 1 receptors causes actin remodelling and myogenic responsiveness in skeletal muscle arterioles , 2016, The Journal of physiology.

[39]  Fei Guo,et al.  Improved prediction of protein-protein interactions using novel negative samples, features, and an ensemble classifier , 2017, Artif. Intell. Medicine.

[40]  Gaotao Shi,et al.  CPPred-RF: A Sequence-based Predictor for Identifying Cell-Penetrating Peptides and Their Uptake Efficiency. , 2017, Journal of proteome research.

[41]  Jijun Tang,et al.  Local-DPP: An improved DNA-binding protein prediction method by exploring local evolutionary information , 2017, Inf. Sci..

[42]  Jijun Tang,et al.  PhosPred-RF: A Novel Sequence-Based Predictor for Phosphorylation Sites Using Sequential Information Only , 2017, IEEE Transactions on NanoBioscience.

[43]  Leyi Wei,et al.  A novel hierarchical selective ensemble classifier with bioinformatics application , 2017, Artif. Intell. Medicine.

[44]  Ran Su,et al.  Identifying N6-methyladenosine sites using multi-interval nucleotide pair position specificity and support vector machine , 2017, Scientific Reports.

[45]  Xiaogang Wang,et al.  T-CNN: Tubelets With Convolutional Neural Networks for Object Detection From Videos , 2016, IEEE Transactions on Circuits and Systems for Video Technology.

[46]  Gaotao Shi,et al.  Fast Prediction of Protein Methylation Sites Using a Sequence-Based Feature Selection Technique , 2019, IEEE/ACM Transactions on Computational Biology and Bioinformatics.