There are over 11000 Structural Genomics (SG) protein structures in the Protein Data Bank. Misannotation rates for SG proteins are estimated to be high since they are annotated simply by their closest sequence or structural match. We have developed a strategy to predict function reliably and to identify these misannotations. Our methods, Th e oretical M icroscopic A nomalous Ti tration C urve S hapes (THEMATICS) and P artial O rder O ptimum L ikelihood (POOL), are used to predict functionally important residues for members of the Six-hairpin glycosidase superfamily. Each functional subclass exhibits a unique local spatial pattern based on the predictions, which are used to sort the superfamily according to biochemical function and also to determine whether SG members of this superfamily are correctly or incorrectly annotated. Using this method, we show that a putative alpha-rhamnosidase from Bacteroides thetaiotaomicron is likely to be correctly annotated, while a putative glucoamylase from Bacteroides fragilis is likely to be incorrectly annotated.

Functional classification of the six-hairpin glycosidase superfamily based on structurally aligned local active sites

244th National Meeting (2012)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.