ContactNet: Geometric-Based Deep Learning Model for Predicting Protein-Protein Interactions
This addresses the problem of identifying accurate protein-protein interaction models for researchers in computational biology, particularly for cases like antibody-antigen interactions where traditional methods like Multiple Sequence Alignment are unavailable, representing a strong specific gain.
The paper tackled the challenge of scoring protein-protein interaction models from docking algorithms, developing ContactNet, an attention-based Graph Neural Network that doubles the accuracy of state-of-the-art scoring functions, achieving 43% Top-10 accuracy for docked antigen-modeled antibody structures and 65% for unbound antibodies.
Deep learning approaches achieved significant progress in predicting protein structures. These methods are often applied to protein-protein interactions (PPIs) yet require Multiple Sequence Alignment (MSA) which is unavailable for various interactions, such as antibody-antigen. Computational docking methods are capable of sampling accurate complex models, but also produce thousands of invalid configurations. The design of scoring functions for identifying accurate models is a long-standing challenge. We develop a novel attention-based Graph Neural Network (GNN), ContactNet, for classifying PPI models obtained from docking algorithms into accurate and incorrect ones. When trained on docked antigen and modeled antibody structures, ContactNet doubles the accuracy of current state-of-the-art scoring functions, achieving accurate models among its Top-10 at 43% of the test cases. When applied to unbound antibodies, its Top-10 accuracy increases to 65%. This performance is achieved without MSA and the approach is applicable to other types of interactions, such as host-pathogens or general PPIs.