Plant & Animal Genomes XVI Conference Plant & Animal Genomes XVI ConferenceJanuary 12-16, 2008
Town & Country Convention Center
San Diego, CA
Igor Grigoriev , Andrea Aerts , Astrid Terry , Alan Kuo , Robert Otillar , Asaf Salamov , Kemin Zhou , Xueling Zhao , Frank Korzeniewski
JGI Annotation pipeline is built for high-throughput annotation of eukaryotic genome assemblies. It includes a suite of gene prediction, annotation, and analysis tools that can be applied to both draft and finished genomes. Gene models are predicted with ab initio predictor FGENESH [Salamov & Solovyev, 2000], specifically trained for each genome, homology-based GENEWISE [Birney & Durbin, 2000] and FGENESH+[Salamov & Solovyev, 2000] and other methods. Available ESTs are used to correct predicted coding regions of the genes and when possible extend them to putatively full-length gene models. Finally, a non-redundant set of gene models is produced, in which a single best gene model per locus is selected on basis of EST support and sequence similarity to known protein sequences.
All predicted genes are annotated according to Gene Ontology (GO) [Ashburner et al, 2000], eukaryotic orthologous groups (KOGs [Koonin et al, 2004]), and KEGG metabolic pathways [Kanehisa et al, 2004]. Predicted genes and annotations are displayed via interactive web-pages on JGI Genome Portal, which provides tools for whole genome analysis and manual curation. Additional experimental data (ESTs, microarrays and proteomics), comparison with other gene predictions and related genomes as well as distributed community-wide manual curation enable further validation and improvement of predicted genes and functions.
Genomes annotated using JGI Annotation pipeline are available at genome.jgi-psf.org.