DeepConv-DTI: Prediction of drug-target interactions via deep learning with convolution on protein sequences

Abstract

Identification of drug-target interactions (DTIs) plays a key role in drug discovery. The high cost and labor-intensive nature of in vitro and in vivo experiments have highlighted the importance of in silico -based DTI prediction approaches. In several computational models, conventional protein descriptors have been shown to not be sufficiently informative to predict accurate DTIs. Thus, in this study, we propose a deep learning based DTI prediction model capturing local residue patterns of proteins participating in DTIs. When we employ a convolutional neural network (CNN) on raw protein sequences, we perform convolution on various lengths of amino acids subsequences to capture local residue patterns of generalized protein classes. We train our model with large-scale DTI information and demonstrate the performance of the proposed model using an independent dataset that is not seen during the training phase. As a result, our model performs better than previous protein descriptor-based models. Also, our model performs better than the recently developed deep learning models for massive prediction of DTIs. By examining pooled convolution results, we confirmed that our model can detect binding sites of proteins for DTIs. In conclusion, our prediction model for detecting local residue patterns of target proteins successfully enriches the protein features of a raw protein sequence, yielding better prediction results than previous approaches. Our code is available at https://github.com/GIST-CSBL/DeepConv-DTI .

In silico Methods for Design of Kinase Inhibitors as Anticancer Drugs

“…Today, many different machine learning methods such as support vector machines (SVM), k-Nearest Neighbors, Artificial Neural Networks (ANN), Deep Learning (DL), etc. are used in pharmaceutical research and they can be applied in various processes of drug design from virtual screening to de novo drug design (Buchwald et al, 2011;Drewry et al, 2017;Konze et al, 2019;Kuthuru et al, 2019;Lee et al, 2019;Zhavoronkov et al, 2019).…”

Dataset Augmentation Allows Deep Learning-Based Virtual Screening To Better Generalize To Unseen Target Classes, And Highlight Important Binding Interactions

“…Each DUD-E target can also be categorized into one of the following families, with 7 numbers indicating the number of targets per category: Kinase (26), Protease (15), Nuclear (11), G protein-coupled receptor (GPCR) (15) and Other (45).…”

Drug-Target Indication Prediction by Integrating End-To-End Learning And Fingerprints

“…Recent studies have also proposed end-to-end compound descriptor learning functions toward ensuring a closer relationship between the learning objective and the input space [11], [15], [16]. Although several studies have approached DTI prediction as a binary classification problem [17], the nature of bioactivity is deemed to be continuous [18]. In [19], a DL model using ECFP (with diameter 4) is compared to a Molecular Graph Convolution (GraphConv) model on predicting binding affinities.…”