Leveraging Local Protein Structures for Enhanced Drug-Target Binding Affinity Predictions Using Deep Learning Techniques

The traditional drug discovery process is both time-consuming and costly. Utilizing artificial intelligence to predict drug-target binding affinity DTA has become a crucial approach for accelerating new drug discovery. This study introduces a novel deep learning-based method that incorporates both the primary and secondary structures of proteins to better represent the local and global features of proteins. We employ convolutional neural networks CNNs and graph neural networks GNNs to model proteins and drugs sep-arately, capturing their interactions more effectively. Our method demon-strated improved performance in predicting DTA compared to state-of-the-art methods on two benchmark datasets.