VADR: validation and annotation of virus sequence submissions to GenBank

Alejandro A Schäffer; Eneida Hatcher; Linda Yankie; Lara Shonkwiler; J Rodney Brister; Ilene Karsch-Mizrachi; Eric P Nawrocki

VADR-based annotation of mature peptides and structural RNA features The central problems in checking and annotating viral genomes and bacterial genomes differ [2] . Bacterial genomes are typically larger with hundreds of genes, many of which may be uncharacterized. Hence, assembly and gene prediction are two of the difficult problems in bacterial annotation. Virus genomes are typically smaller with only a handful of genes, but the biological phenomena of cleavage of proteins into mature peptides [24] and ribosomal slippage [25, 26, 27] are prominent. Surprisingly, while 97% (31,086 of 32, 190) and 85% (17,900 of 20,973) of non-RefSeq norovirus and dengue virus sequences, respectively, last modified between January 1, 1989 and December 31, 2017, had at least one CDS annotated, only 1.4% (443 of 31,086) and 6.4% (1,157 of 17,900) of those those had any mature peptides annotated, even though mature peptide cleavage occurs in all viruses of the corresponding genera. VADR fixes this omission; all norovirus and dengue virus sequences in GenBank annotated with VADR have mature peptides predicted, where possible.