Improving Prediction of Drug-Target Binding Affinity using Attention Mechanism and Bi-Directional Long Short-Term Memory
Min Gao, Shaohua Jiang, Zhijian Lyu, Weibin Ding, Ting Xu · 2023
The successful identification of drug-target interactions (DTIs) plays an important role in the drug discovery process and drug repurposing. However, most methods for DTIs prediction ignore a continuous value, namely binding affinity, and it is an essential piece of information about protein-ligand interactions. In this study, we propose a deep learn-based prediction model called HABiLSTM-DTA, which is designed for prediction of drug-target affinity. The main innovation is the application of the attention mechanism to model the complex interactions between drug molecules and protein sequences. We also use convolutional layers and bi-directional long short-term memory to learn their feature representations. We evaluate the proposed model on two benchmark datasets. The results show that the proposed model achieves relatively better results than other 1D sequence-based baselines, and it is an effective method for predicting drug target binding affinity.