Bovidae Genome Data at NCBI – from Genome Annotation to Gene Curation

The National Center for Biotechnology Information (NCBI) provides annotated Bovidae genomes, genes, transcripts, and proteins in the reference sequence (RefSeq) database. This content is generated using NCBI’s Eukaryotic Genome Annotation Pipeline which leverages evidence-based computational prediction and manually curated data. RefSeq prioritizes supporting more complete genome assemblies and aims to provide genome annotation for, typically, one reference genome per species although we may elect to annotate alternate assemblies, as has been the case for Bos taurus. The core of our evidence-based annotation pipeline generates gene, transcript, and protein models based on alignment analysis of: transcript data available in GenBank, RNAseq datasets available in the Sequence Read Archive (SRA), curated RefSeq transcripts and proteins from Bovidae and human, and UniProtKB/Swiss-Prot records. Annotation products are accessible through several NCBI resources, such as Gene, Nucleotide, Protein, BLAST and the (new) Genome Data Viewer. Manual curation is performed at both gene and sequence level and includes evidence-based generation of RefSeq transcripts and proteins, correction of sequence errors and improvement of gene annotation based on quality assurance tests. Advantages of the RefSeq product include: consistent annotation and curation protocols; public availability of the data in multiple resources and formats; clear data provenance; and the ability to add, remove, or update transcript and protein products between whole genome annotation cycles. This presentation will provide an overview of RefSeq data processing, Bovidae genome resources, and 2016 annotation updates of Bos taurus and Capra hircus.