Deep learning applications for predicting pharmacological properties of drugs and drug repurposing using transcriptomic data

By Alexander M. Aliper, Sergey M. Plis, Artem V. Artemov, Alvaro E. Ulloa, Polina Mamoshina, A. Zhavoronkov. 2016

495 citations recorded by the source corpus.

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 dataset of samples perturbed with different concentrations of the drug for 6 and 24 hours. In both gene and pathway level classification, DNN convincingly outperformed support vector machine (SVM) model on every multiclass classification problem, however, models based on a pathway level classification perform 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.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Opportunities and obstacles for deep learning in biology and medicine (2017, methodology): “[35] used gene- and pathway-level drug perturbation transcriptional profiles from the Library of Network-Based Cellular Signatures [288] to train a fully connected deep neural network to predict drug therapeutic uses and indications.”
Artificial Intelligence, Machine Learning, Deep Learning, and Cognitive Computing: What Do These Terms Mean and How Will They Impact Health Care? (2018, background): “algorithms that accurately identify drug toxicity without the usual decade of animal testing [21] 242 In fact, they encourage it.”
Artificial intelligence in drug development (2025, unknown): “246), we can be judiciously optimistic about AI’s promise for accelerating drug development and benefiting human health.”
Machine learning applications in drug development (2019, background): “ML techniques are now extremely popular in drug development (see [13,27,31] for recent surveys) as they allow automation of highly-dimensional, noisy biological data analysis.”
Molecular Sets (MOSES): A Benchmarking Platform for Molecular Generation Models (2018, unknown): “The exponential growth of publicly available experimental data also increases popularity of machine learning applications in bioinformatics [9, 10] for predicting pharmacological properties of drugs [11], biomarker development [12–15], and target identiﬁcation [16].”
AI in Medical Imaging Informatics: Current Challenges and Future Directions (2020, unknown): “Undoubtedly, research in deep learning applications and methods is expected to grow, especially in in view of documented advances across the spectrum of healthcare data, including EHR [260], genomic [261], [262], physiological parameters [263], and natural language data processing [264].”
Rethinking drug design in the artificial intelligence era (2019, methodology): “For instance, artificial neural networks are being employed to classify complex cellular phenotypes and build predictive models that are used for drug repurposing 52,53 .”
Automating drug discovery (2017, background): “Several heterogeneous deep-learning systems with high prediction accuracies have been developed for drug–target association, drug repurposing opportunities and target identification, among other tasks 202,206,207,208 . and software 180,181 .”
The rise of deep learning in drug discovery. (2018, unknown): “Aliper et al. [43] built DNN models for predicting pharmacological properties of drugs and for drug repurposing leveraging transcriptomic data from the LINCS project [44], as well as the pathway information.”
The cornucopia of meaningful leads: Applying deep adversarial autoencoders for new molecule development in oncology (2016, background): “…of this flexibility and adaptability of DNN for learning from large range of data, DNNs are now considered as an increasingly important area in the biomedical field that shows significant potential in comprehensive -omics analysis and could be useful for tackling many current issues [1, 8, 9].”
Artificial intelligence in drug development: present status and future prospects. (2019, background): “In a study performed by Aliper et al ., it was demonstrated that DNNs could classify complex drug action mechanisms on the pathway level, thus classifying drugs into therapeutic categories according to their functional class, efﬁ-cacy, therapeutic use and toxicity [61].”
Converging blockchain and next-generation artificial intelligence technologies to decentralize and accelerate biomedical research and healthcare (2015, methodology): “For example, Aliper and colleagues used signaling pathway analysis to reduce the dimensionality of drug-induced gene expression profiles and to train a DNN based predictor of pharmacological properties of drugs [33].”
druGAN: An Advanced Generative Adversarial Autoencoder Model for de Novo Generation of New Molecules with Desired Molecular Properties in Silico. (2017, background): “[7, 8], drug discovery [4, 9, 10], predicting clinical trial outcomes [11], and generating”