Efficient Similarity Search in Protein Structure Databases: Improving Cliqae-Detection through Clique Hashing.

摘要: In order to make the structural comparison of protein binding sites more efficient, we propose a two-step method that combines advantages from both graph-based clique-detection and geometric hashing. The search for similarity is completely independent sequence fold information. Instead, it based on recent approach automatic extraction structures representation their physicochemical properties. We also present some empirical results in medium-sized dataset.