Progress in predicting protein function from structure: unique features of O-glycosidases.

Area of Science:

• Structural biology
• Enzymology
• Bioinformatics

Background:

• The Structural Genomics Initiative will generate numerous new protein structures.
• Predicting protein function from structure is crucial, especially for proteins with low sequence similarity to known ones.
• O-glycosidases, which cleave O-glycosidic bonds, represent a diverse group of enzymes with independent evolutionary origins and varied folds.

Purpose of the Study:

• To identify conserved structural characteristics of O-glycosidases that are independent of sequence and fold.
• To investigate the utility of electrostatic surfaces in distinguishing O-glycosidase function.
• To demonstrate the feasibility of predicting O-glycosidase function directly from protein structure.

Main Methods:

• Analysis of unique structural features across diverse O-glycosidase families.
• Examination of electrostatic surface properties of O-glycosidases.
• Structure-based function prediction methodologies.

Main Results:

• O-glycosidases exhibit distinctive structural features that transcend sequence and fold classifications.
• Electrostatic surfaces are particularly characteristic and informative for this enzyme class.
• Accurate prediction of O-glycosidase function from structural data is achievable.

Conclusions:

• Conserved structural motifs, particularly electrostatic surface properties, can be leveraged for function prediction in enzyme superfamilies.
• This approach is vital for annotating newly determined protein structures from large-scale genomics projects.
• Structure-based functional prediction offers a powerful complement to sequence-based methods for uncharacterized proteins.