A Graph Method for Predicting Protein-RNA Interaction Using Reduced Amino Acid Alphabets

Protein-RNA interactions play important roles in the biological systems. Searching for regular patterns at the protein-RNA binding interfaces is important for understanding how protein and RNA recognize each other and bind to form a complex. Herein we present a graph mining approach to discovering patterns at the protein-RNA interfaces. We represented known protein-RNA interfaces using graph and then discovered graph patterns enriched in the interfaces. To evaluate the effectiveness of our method, we conducted two experiments. The first one was to use the graph patterns as input features to a Support Vector Machine to automatically classify protein surface patches into RNA-binding sites and non-RNA-binding sites. In the second experiment, we built a simple scoring function that calculated the total number of the graph patterns that occurred in a protein-RNA interface. That scoring function was able to discriminate near native protein-RNA complexes from docking decoys with a performance comparable with a state-of-the-art complex scoring function. Our work also compared the performances of the proposed method using full amino acid alphabet with reduced alphabets. The results indicate that our method yields strikingly good result in terms of hit count.