| Hi-index | 0.00 |
Recently, several domain based computational models for predicting protein-protein interactions (PPIs) have been proposed. However, the majority of these models often have limitations in providing detailed information on which domain pair (single domain interaction) or DC pair (multi domain interaction) will actually interact for a predicted protein interaction. To solve this, we developed a computational model to predict PPIs using information of intraprotein domain cohesions and inter-protein domain or DC coupling interactions. A method to identify the primary interacting DC pair is also incorporated into the model for inferring actual participants in a predicted interaction. Our method made a significant improvement in PPI prediction accuracy, and the primary interacting DC pair identification turned out valid specifically in predicting multi domain protein interactions.