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

As of September 2019, GenBank [1] contained more than 3 million viral sequences totaling over 4 billion nucleotides in length and including over 180,000 complete genomes for viruses other than influenza. More than 250,000 of these sequences were submitted in 2018. All sequence submissions are validated prior to deposition in GenBank. Automated validation and annotation methods become increasingly important as sequence submission numbers grow. Table 1 shows the number of sequences for the 16 virus species with the most sequences in GenBank. Influenza sequences are the second most abundant and the National Center of Biotechnology Information (NCBI), where GenBank is housed, has expended considerable effort to organize flu sequences and streamline the submission of new influenza virus sequences, including a tool to validate and annotate flu submissions called FLAN [2] . The influenza virus sequence submission tool (https://www.ncbi.nlm.nih.gov/ genome/viruses/variation/help/flu-help-center/submitflu-sequences/) is implemented specifically for influenza with many hard-coded features. It has proven [3] (https:// www.ncbi.nlm.nih.gov/genome/viruses/variation) includes specialized components that attempt to normalize annotation of previously submitted sequences for rotaviruses, dengue virus, West Nile virus, ebolaviruses, Zika virus, and MERS coronavirus. The resulting standardized annotation supports virus-specific searches using gene and protein names. However, the Virus Variation Resource does not support the submission of new sequences, and its tools are not generalizable, such that creating components for additional virus species is laborious.