Revising the way we conceive and name viruses below the species level: a review of geminivirus taxonomy calls for new standardized isolate descriptors.

Geminivirus taxonomy and nomenclature is increasing in complexity with time, and the growing number of geminivirus sequences deposited in gene banks requires regular taxonomic updates and calls for new descriptors to identify virus isolates unambiguously. Fauquet et al. [1] proposed a system to standardize the names of the viruses, and corresponding guidelines have been followed since, rendering nomenclature much easier. Recently, due to difficulties inherent in species identification, the ICTV Geminiviridae Study Group proposed new species demarcation criteria, the most important of which being an 89% identity threshold between complete DNA-A component nucleotide sequences of begomoviruses. This threshold has been utilised since with general satisfaction. In this paper, we review the status of geminivirus species demarcation and nomenclature for a total of 389 isolates. A small number of corrections have been made to comply with the adopted demarcation criteria but otherwise the classification system has remained robust and therefore we propose to continue using it. However, the large numbers of geminivirus sequences that have become available have led us to recognize the need for a better description of virus isolates. The pairwise comparison distribution below the taxonomic level of species identified two peaks, one at 90-91% identity that may correspond to "strains" and one at 96-98% identity that may correspond to "variants". Guidelines for descriptors for each of these levels are proposed to standardize nomenclature. As a consequence, we have revisited the status of some virus isolates to elevate them to "strains". An updated list of all geminivirus isolates currently available is provided.