Protein-Ligand Binding Affinity Prediction Based on Deep Learning

The identification of novel drug target (DT) interactions is an important part of the drug discovery process. A large number of studies have investigated whether DT interacts through dichotomies, yet the strength of the interaction between ligand and protein can be imagined as a continuous value of binding affinity. At present, many methods have been proposed to predict this value, most of which need to determine the three-dimensional structure of proteins, but the structure of some proteins is difficult to know. In this paper, we propose a deep learning-based approach to predict binding affinity that does not rely on three-dimensional structure, but instead takes proteins and their structural properties and ligand sequences as input features. Compared to other methods that utilize the 3D structural characteristics of proteins, this model exhibits better performance.