GNAS2009 Abstract #5 - Segerdell, Erik

Title:
Xenopus gene nomenclature in Xenbase

Authors:
Erik Segerdell, Kevin Snyder and Peter Vize

Affiliation:
Department of Biological Sciences, University of Calgary, 2500 University Drive NW, Calgary, Alberta, Canada

Abstract:

The Xenopus gene catalog is at the core of Xenbase (www.xenbase.org), the model organism database for the African frogs X. laevis and tropicalis. The catalog currently has 4,705 manually annotated gene pages and 10,833 machine annotated gene pages. Of the 4,705 manually curated genes, there are 4,159 with synteny to other organisms determined via Metazome, our primary method for establishing orthology. Another 546 manually annotated genes have sequence similarity with another organism but lack synteny evidence or are novel Xenopus genes. Xenbase uses mammalian nomenclature to facilitate accessibility to the broader research community and approved Xenopus gene names are identical to human gene names whenever possible. Manually annotated genes lacking synteny evidence are displayed as 'provisional'. The 10,833 machine-annotated catalog entries are marked as 'provisional - machine annotated only'; of these, 9,264 entries were added based on sequence similarity to a human gene based on a maximum e-value of 1e-10 with a minimum 55% identity and 65% coverage. Their symbols and names are displayed with the stipulation that they were machine predicted matches using Entrez gene. The remaining 1,558 genes were added based on synteny evidence to some other organism. Their symbols are set as 'unnamed' and their names are set as the Metazome homology description.

To ensure approved Xenopus names stay synchronized with ever-changing HGNC names, we use a script system that checks Xenbase names for an exact match against the associated human Entrez Gene name. When a mismatch is detected an email is automatically generated notifying a curator, who revisits the Xenopus gene and if necessary updates its name.