Deep learning and virtual drug screening. (5th October 2018)

• Record Type: Journal Article
• Authors: Carpenter, Kristy A; Cohen, David S; Jarrell, Juliet T; Huang, Xudong
• Is Part Of: Future medicinal chemistry. Volume 10:Number 21(2018)
• Page Start: 2557
• Page End: 2567
• Publication Date: 2018-10-05
• Subjects: artificial intelligence -- artificial neural networks -- convolutional neural networks -- deep learning -- drug discovery -- machine learning -- multitask learning -- virtual screening; Pharmaceutical chemistry -- Periodicals; 615.19005
• DOI: 10.4155/fmc-2018-0314
• Languages: English
• ISSNs: 1756-8919

Abstract

Current drug development is still costly and slow given tremendous technological advancements in drug discovery and medicinal chemistry. Using machine learning (ML) to virtually screen compound libraries promises to fix this for generating drug leads more efficiently and accurately. Herein, we explain the broad basics and integration of both virtual screening (VS) and ML. We then discuss artificial neural networks (ANNs) and their usage for VS. The ANN is emerging as the dominant classifier for ML in general, and has proven its utility for both structure-based and ligand-based VS. Techniques such as dropout, multitask learning and convolution improve the performance of ANNs and enable them to take on chemical meaning when learning about the drug-target-binding activity of compounds. Graphic abstract: