YAPS: Yet Another Protein Similarity

In this article we present a novel method for measuring protein similarity based on their tertiary structure. Our new method deals with suffix trees and classical information retrieval tasks, such as the vector space model, using tf-idf term weighing schema or using various types of similarity measures. Our goal to use the whole PDB database of known proteins, not just some kinds of selections, which have been studied in other works. For verification of our algorithm we are using comparisons with the SCOP database which is maintained primarily by humans. The next goal is to be able to categorize proteins not included in the latest version of the SCOP database with nearly 100% accuracy.