Dataset Information

Deep Learning Applications for Predicting Pharmacological Properties of Drugs and Drug Repurposing Using Transcriptomic Data.

ABSTRACT: Deep learning is rapidly advancing many areas of science and technology with multiple success stories in image, text, voice and video recognition, robotics, and autonomous driving. In this paper we demonstrate how deep neural networks (DNN) trained on large transcriptional response data sets can classify various drugs to therapeutic categories solely based on their transcriptional profiles. We used the perturbation samples of 678 drugs across A549, MCF-7, and PC-3 cell lines from the LINCS Project and linked those to 12 therapeutic use categories derived from MeSH. To train the DNN, we utilized both gene level transcriptomic data and transcriptomic data processed using a pathway activation scoring algorithm, for a pooled data set of samples perturbed with different concentrations of the drug for 6 and 24 hours. In both pathway and gene level classification, DNN achieved high classification accuracy and convincingly outperformed the support vector machine (SVM) model on every multiclass classification problem, however, models based on pathway level data performed significantly better. For the first time we demonstrate a deep learning neural net trained on transcriptomic data to recognize pharmacological properties of multiple drugs across different biological systems and conditions. We also propose using deep neural net confusion matrices for drug repositioning. This work is a proof of principle for applying deep learning to drug discovery and development.

SUBMITTER: Aliper A

PROVIDER: S-EPMC4965264 | biostudies-literature | 2016 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Deep Learning Applications for Predicting Pharmacological Properties of Drugs and Drug Repurposing Using Transcriptomic Data.

Aliper Alexander A Plis Sergey S Artemov Artem A Ulloa Alvaro A Mamoshina Polina P Zhavoronkov Alex A

Molecular pharmaceutics 20160608 7

Deep learning is rapidly advancing many areas of science and technology with multiple success stories in image, text, voice and video recognition, robotics, and autonomous driving. In this paper we demonstrate how deep neural networks (DNN) trained on large transcriptional response data sets can classify various drugs to therapeutic categories solely based on their transcriptional profiles. We used the perturbation samples of 678 drugs across A549, MCF-7, and PC-3 cell lines from the LINCS Proje ...[more]

PMID: 27200455

Similar Datasets

Project description:Drug repositioning or repurposing, i.e. identifying new indications for existing drugs, has gained increasing attention in the recent years. This approach enables the scientists to discover "new targets" for known drugs in a cost and time efficient manner. Glycation, the non-enzymatic reaction of sugars with proteins or nucleic acids to form early glycation (Amadori or fructosamine) products, is a key molecular basis of diabetic complications. Inhibiting the process of non-enzymatic protein glycation is one of the key strategies to prevent glycation-mediated diabetic complications. The present study focuses on the anti-glycation activity of 18 drugs, commonly used for the treatment of gastrointestinal, central nervous system, inflammatory diseases, bacterial infections, and gout. This study was carried out by using two in-vitro protein anti-glycation assay models. Results revealed that nimesulide (3), a non-steroidal anti-inflammatory drug, possesses a good anti-glycation activity in in-vitro BSA-MG and BSA-glucose glycation models with IC50 values of 330.56 ± 2.90, and 145.46 ± 16.35 μM, respectively. Phloroglucinol dihydrate (11), a drug used for the treatment of gastrointestinal diseases, showed a weak activity in BSA-MG glycation model (IC50 = 654.89 ± 2.50 μM), while it showed a good activity in BSA-glucose assay (IC50 = 148.23 ± 0.15 μM). Trimethylphloroglucinol (9), a drug used for the treatment of pain related to functional disorders of the digestive and biliary tracts, also showed a good antiglycation activity in BSA-MG model (IC50 = 321.15 ± 1.26 μM), while it was found to be inactive in in-vitro BSA-glucose assay (IC50 = 12.95% inhibition). These activities of drugs were compared with the anti-glycation activity of the standard, rutin (IC50 = 294.5 ± 1.50 μM in BSA-MG glycation model, and IC50 = 86.94 ± 0.24 μM in BSA- glucose model). Rest of the drugs exhibited a relatively weak antiglycation activity. This study identifies nimesulide (3), and phloroglucinol dihydrate (11) as new inhibitors of in-vitro protein glycation for further investigations as potential anti-diabetic agents.

Project description:MotivationWhile drug combination therapies are a well-established concept in cancer treatment, identifying novel synergistic combinations is challenging due to the size of combinatorial space. However, computational approaches have emerged as a time- and cost-efficient way to prioritize combinations to test, based on recently available large-scale combination screening data. Recently, Deep Learning has had an impact in many research areas by achieving new state-of-the-art model performance. However, Deep Learning has not yet been applied to drug synergy prediction, which is the approach we present here, termed DeepSynergy. DeepSynergy uses chemical and genomic information as input information, a normalization strategy to account for input data heterogeneity, and conical layers to model drug synergies.ResultsDeepSynergy was compared to other machine learning methods such as Gradient Boosting Machines, Random Forests, Support Vector Machines and Elastic Nets on the largest publicly available synergy dataset with respect to mean squared error. DeepSynergy significantly outperformed the other methods with an improvement of 7.2% over the second best method at the prediction of novel drug combinations within the space of explored drugs and cell lines. At this task, the mean Pearson correlation coefficient between the measured and the predicted values of DeepSynergy was 0.73. Applying DeepSynergy for classification of these novel drug combinations resulted in a high predictive performance of an AUC of 0.90. Furthermore, we found that all compared methods exhibit low predictive performance when extrapolating to unexplored drugs or cell lines, which we suggest is due to limitations in the size and diversity of the dataset. We envision that DeepSynergy could be a valuable tool for selecting novel synergistic drug combinations.Availability and implementationDeepSynergy is available via www.bioinf.jku.at/software/DeepSynergy.Contactklambauer@bioinf.jku.at.Supplementary informationSupplementary data are available at Bioinformatics online.

Dataset Information

Deep Learning Applications for Predicting Pharmacological Properties of Drugs and Drug Repurposing Using Transcriptomic Data.

Publications

Deep Learning Applications for Predicting Pharmacological Properties of Drugs and Drug Repurposing Using Transcriptomic Data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets