Comprehensive Survey of Recent Drug Discovery Using Deep Learning.

Jintae Kim,Wankyu Kim,Dongbo Min,Sera Park

doi:10.3390/ijms22189983

Abstract

Drug discovery based on artificial intelligence has been in the spotlight recently as it significantly reduces the time and cost required for developing novel drugs. With the advancement of deep learning (DL) technology and the growth of drug-related data, numerous deep-learning-based methodologies are emerging at all steps of drug development processes. In particular, pharmaceutical chemists have faced significant issues with regard to selecting and designing potential drugs for a target of interest to enter preclinical testing. The two major challenges are prediction of interactions between drugs and druggable targets and generation of novel molecular structures suitable for a target of interest. Therefore, we reviewed recent deep-learning applications in drug–target interaction (DTI) prediction and de novo drug design. In addition, we introduce a comprehensive summary of a variety of drug and protein representations, DL models, and commonly used benchmark datasets or tools for model training and testing. Finally, we present the remaining challenges for the promising future of DL-based DTI prediction and de novo drug design.

Highlights

The primary goal of drug discovery is to develop safe and effective medicines for human diseases
DUD-E [166], a gold standard dataset used for the evaluation of the virtual screening (VS) methods, selected decoys based on the concept that the decoy compounds must be structurally different from the known ligands to reduce the false negative, whereas it must be similar to the known ligands with respect to physicochemical properties to reduce bias
The drug–target interaction (DTI) prediction can be grouped into two types: (1) drug–target pairs (DTPs) prediction by a classification model that assigns a positive or negative label to the DTP and (2) drug–target affinity (DTA) prediction by a regression model that estimates the binding affinity value between the drug and target

Summary

Introduction

The primary goal of drug discovery is to develop safe and effective medicines for human diseases. All the drug development processes—from target identification to step-bystep clinical trials—require significant amount of time and cost. Computer-aided drug discovery or design methods have played a major role in this hit-to-lead process by reducing the burden of consumptive validation experiments since the 1980s in modern pharmaceutical research and development (R & D) [2,3,4]. Even this in silico approach has not prevented the decline in pharmaceutical industry R & D productivity since the mid-1990s

Objectives

Methods

Results