RefSeq Annotation and Curation of the Tomato Genome

NCBI’s Reference Sequence (RefSeq) project is a publicly available database of annotated genomic, transcript, and protein sequence records. RefSeq records for plants, in most part, are generated using NCBI’s eukaryotic genome annotation pipeline. The annotation process involves prediction of genes, transcripts and proteins (XM_, XP_ and XR_ accessions) for finished and unfinished genome assemblies submitted to GenBank. A small subset of “known” records (NM_, NP_, NR_ accessions) are generated and maintained independently of the annotation process by a combination of automated processing and manual review.

Currently 35 plant genomes have been annotated by NCBI, including four Solanaceae genomes: tomato, potato and two tobacco species, Nicotiana sylvestris, and Nicotiana tomentosiformis. The tomato genome was chosen for manual curation because of its importance to the plant community and the availability of large-scale transcript data. This transcript-based curation process of “known” dataset focuses on improving the quality of the represented sequence by assessing indels and mismatches among the RefSeq transcripts, removing chimeric and redundant transcripts and genes, updating product names and creating alternatively spliced transcript variants and protein isoforms. The curation process also allows for the extension of UTRs to have the representation of full gene. This curated dataset defines the ‘best’ sequence representation which, in turn, is used as a primary input reagent for annotating other plant genomes, of the same or related family.

The plant genomes annotated by this pipeline are reported publicly with links to FTP, BLAST, MapViewer, and a detailed Annotation Report Summary at http://www.ncbi.nlm.nih.gov/genome/annotation_euk/all/.

RefSeq data can be accessed at the RefSeq homepage http://www.ncbi.nlm.nih.gov/refseq/ or can be downloaded from the FTP directory at ftp://ftp.ncbi.nlm.nih.gov/genomes/refseq/plant/Solanum_lycopersicum/.