Abstract
The rapid spread of novel coronavirus pneumonia (COVID-19) has led to a dramatically increased mortality rate worldwide. Despite many efforts, the rapid development of an effective vaccine for this novel virus will take considerable time and relies on the identification of drug-target (DT) interactions utilizing commercially available medication to identify potential inhibitors. Motivated by this, we propose a new framework, called DeepH-DTA, for predicting DT binding affinities for heterogeneous drugs. We propose a heterogeneous graph attention (HGAT) model to learn topological information of compound molecules and bidirectional ConvLSTM layers for modeling spatio-sequential information in simplified molecular-input line-entry system (SMILES) sequences of drug data. For protein sequences, we propose a squeezed-excited dense convolutional network for learning hidden representations within amino acid sequences; while utilizing advanced embedding techniques for encoding both kinds of input sequences. The performance of DeepH-DTA is evaluated through extensive experiments against cutting-edge approaches utilising two public datasets (Davis, and KIBA) which comprise eclectic samples of the kinase protein family and the pertinent inhibitors. DeepH-DTA attains the highest Concordance Index (CI) of 0.924 and 0.927 and also achieved a mean square error (MSE) of 0.195 and 0.111 on the Davis and KIBA datasets respectively. Moreover, a study using FDA-approved drugs from the Drug Bank database is performed using DeepH-DTA to predict the affinity scores of drugs against SARS-CoV-2 amino acid sequences, and the results show that that the model can predict some of the SARS-Cov-2 inhibitors that have been recently approved in many clinical studies.
Highlights
One of the preliminary stages of drug discovery is the determination of innovative candidate drug compounds that interact with particular target proteins
We propose a novel drug-target interactions (DTI) prediction framework that utilizes a heterogeneous graph attention (HGAT) [23] for efficient modeling of interactions of various targeted topological representation of drugs
Our model primarily comprises four major blocks: (1) the first upper module is introduced to learn protein structure representation using Dense Net augmented with Squeeze and Excite (SE) operation; (2) simultaneously, a heterogeneous graph network is introduced to learn the topological representation of drug molecules; (3) the sequential characteristics of simplified molecular-input line-entry system (SMILES) representation of input compounds is learned through bidirectional ConvLSTM architecture; and (4) the extracted representation is concatenated and fed into the output layer for affinity score calculation
Summary
One of the preliminary stages of drug discovery is the determination of innovative candidate drug compounds that interact with particular target proteins. Through in vivo and in vitro studies, several high-throughput experiments have been conducted to identify the novel compounds with the anticipated interactive characteristics [1]. Expensive costs and chronological order requirements make it impracticable to scan immense volumes of targets. The recent number of FDA-approved drugs is about 10000, according to DrugBank [5]. Only a small number of proteins in the human proteome are targeted by recognized drugs. Knowledge of the drug–target (DT) space is still incomplete and requires a novel approach to enable broader investigation [8].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
