Resources and Tools for Correcting Misannotation of Enzymes

University Of California-San Francisco, San Francisco CA

Investigators

Abstract

The University of Califorina - San Francisco is awarded a grant develop resources and tools for identifying and repairing misannotations of molecular functions for the huge numbers of enzymes discovered in the genome projects and represented in public databases. The project extends previously developed theory and computational methods that have been used to describe superfamilies of enzymes whose structures and sequences look similar but whose functions can be very different, leading to high levels of misannotation when these structural similarities are used to infer similarity of function. This project will produce an ontology, e.g., rules that can be applied generally, describing misannotations, as well as software tools for updating annotations, a reference database of known misanotations, and online training materials to assist end users and database curators in handling misannotations. The ontology, database, and software tools will be integrated into an existing Structure-Function Linkage Database developed and maintained by the PI at the University of California - San Francisco. Creation of a general ontology of misannotation of enzymes will provide a broadly applicable resource for assuring the quality of functional annotations in public sequence and structural databases. For experimental biologists especially, this work will provide resources, stand-alone tools, and online tutorials to help them identify problems in annotations of enzymes with which they work and correct those problems to improve the overall quality of their research.