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

To understand the operational semantics we imposed on VAPiD and VIGOR, it is necessary to begin with the point that they were designed to search the submitted sequence against one or more large databases of known sequences. These databases play a role analogous to the RefSeqs for VADR in that the best match in the database to the input sequence defines what features should be annotated. Because our tests used real sequences already in GenBank and the VAPiD and VIGOR databases also contain sequences already in GenBank, there turns out to be a non-negligible probability that our query sequences are already "known to" VAPiD and/or VIGOR.