Remote Homology Detection Using a Kernel Method that Combines Sequence and Secondary-Structure Similarity Scores

摘要

Distant evolutionary relationships between proteins with low sequence similarity are difficult to recognise bycomputational methods. Consequently, many sequences obtained from large-scale sequencing projects cannot be assigned toany known proteins or families despite being evolutionarily related. To boost sensitivity, various sequence-based methodshave been modified to make use of the better conserved secondary structure. Most of these methods are instance-based orgenerative. Here, we introduce a kernel-based remote homology detection method that allows for a combination of sequenceand secondary-structure similarity scores in a discriminative approach. We studied the ability of the method to predict superfamily membership as defined by the SCOP database. We show that akernel method that combined sequence similarity scores with predicted secondary-structure similarity scores performed similarto a classifier that used scores calculated from sequences and true secondary structures, but performed better than a sequence-onlybased classifier and achieved a better mean than recently published results on the same data-set. It can be concluded that SVM classifiers trained to predict homology between distantly related proteins, become moreaccurate, if a joint sequence/secondary-structure similarity score approach is used.