CGBVS‐DNN: Prediction of Compound‐protein Interactions Based on Deep Learning

Abstract

Computational prediction of compound-protein interactions (CPIs) is of great importance for drug design as the first step in in-silico screening. We previously proposed chemical genomics-based virtual screening (CGBVS), which predicts CPIs by using a support vector machine (SVM). However, the CGBVS has problems when training using more than a million datasets of CPIs since SVMs require an exponential increase in the calculation time and computer memory. To solve this problem, we propose the CGBVS-DNN, in which we use deep neural networks, a kind of deep learning technique, instead of the SVM. Deep learning does not require learning all input data at once because the network can be trained with small mini-batches. Experimental results show that the CGBVS-DNN outperformed the original CGBVS with a quarter million CPIs. Results of cross-validation show that the accuracy of the CGBVS-DNN reaches up to 98.2 % (σ<0.01) with 4 million CPIs.

Novel pathways of HIV latency reactivation revealed by integrated analysis of transcriptome and target profile of bryostatin

“…Prediction of the targets of latency-reversal agent bryostatin. In order to understand the action mechanism of LRA bryostatin, two methods of CGBVS 54,55 and 3NN 56 were used to predict the targets of bryostatin. The CGBVS is a chemical genomics-based virtual screening method for target prediction 54 .…”

“…In order to understand the action mechanism of LRA bryostatin, two methods of CGBVS 54,55 and 3NN 56 were used to predict the targets of bryostatin. The CGBVS is a chemical genomics-based virtual screening method for target prediction 54 . CGBVS predicts compound-protein interactions (CPIs) by using machine learn only based protein sequence rather than three-dimensional structures.…”

Deep learning: from chemoinformatics to precision medicine

“…In the area of pharmaceutical discovery and development, chemoinformatics studies have used many machine learning methods such as decision trees, support vector machines, k-nearest neighbors, and naïve Bayesian classifiers (Mitchell 2014). A prediction method for compound-protein interactions from descriptors of ligands and proteins that use a deep learning technique called "Chemical Genomics-Based Virtual Screening-Deep Neural Networks (CGBVS-DNN)" has been proposed by (Hamanaka et al 2016). At the Merck Molecular Activity Kaggle Challenge, a deep neural network model for analyzing large QSAR datasets won first prize .…”

GANsDTA: Predicting Drug-Target Binding Affinity Using GANs

“…Deep learning can represent and recognize the hidden patterns in the data well, therefore, deep-learning based methods have been proposed to predict DTI or DTA utilizing deep neural networks (DNN) (Peng-Wei et al, 2016;Tian et al, 2016;Hamanaka et al, 2017), convolutional neural networks(CNN), (Jastrzebski et al, 2016;Gomez-Bombarelli et al, 2018) recurrent neural networks (RNNs) and stacked-autoencoders based architectures. These methods facilitate the learning of the 3D structures provided and the bimolecular interaction mechanism.…”