Annotation and comparative analysis

1. Annotation is performed by viewing the output of a composite of software tools, both homology-based and de-novo. Annotation cannot be completely verified, but the veracity is increased with the amount of tools used. Homology databases include Unigene, EST for both the species and other species, mus trace database, protein database and the nt database that includes cDNAs. Two de-novo genefinders are used that were found to have the most accuracy: genscan and fgenesH. Genes are not annotated unless they have either EST, cDNA or protein homology available. A description can be found at http://pga.lbl.gov/annot_descr.html

2. Comparative analysis of genomic sequences of different species is performed by the VISTA software package that includes a global alignment program AVID and a visualization module with multiple annotation and display options. The programs have been extensively tested and debugged since we released them in the fall of 2000. The public VISTA server http://www-gsd.lbl.gov/vista processed about 5500 queries, and more than 300 copies were distributed in 26 countries. We are constantly working on further improvements of the tools.