Cross-protection is a virus-virus antagonistic interaction well-known from plant virologists as it has been used for decades to prevent viral diseases for several plant species. It consists in infecting a plant with a primary virus in order to prevent (or interfere with) its subsequent infection by a genetically related virus1. A key parameter of this phenomenon, highlighted in almost all publications dealing with this phenomenon, is that it works only between genetically closely related viruses. Surprisingly, despite the importance of this feature in cross-protection, the degree of sequence identity required for cross-protection to occur has not been clearly determined. The information is usually limited to the fact that cross-protection occurs between viruses belonging to the same species but not between different species. However, for Pepino mosaic virus (PepMV) and Citrus tristeza virus (CTV), it was further shown that cross-protection occurs only between variants belonging to the same but not different strains2,3. It is interesting to note that the definition of a strain differs for these two viral species, with a 95% global nucleotide sequence identity threshold for PepMV and a definition based on the ORF1a phylogeny for CTV with sequence identity ranging from 72.1% to 91.2 %2–4. While the genetic relatedness required between protecting and challenging viruses might differ depending on the pathosystem, these findings raise questions concerning (i) the sequence identity required for cross-protection success and (ii) whether this identity is global or partial, i.e. whether it corresponds to the whole genome or only portion(s) of it. In other words, while genetic relatedness appears to be a decisive feature of cross-protection, how closely related do protecting and challenging variants need to be is still unknown. The question of the genetic relatedness between variants required for cross-protection success is all the more puzzling for multipartite viruses, which usually share limited sequence identity between their different segments, thus raising questions about the way this phenomenon works for these viruses. Grapevine fanleaf virus (GFLV) is a bipartite virus considered as one of the most damaging viruses for vines. By using (i) a collection of natural GFLV challenging variants with decreasing sequence identity to a primary variant and spanning the known GFLV genetic diversity, and (ii) recombinant viruses, for which large or small portions of the genomes have been swapped between distinct strains, we are studying whether cross-protection is based on a global or a partial sequence identity. Our approach should further enable us to calculate the percentage of identity required and thus identify which GFLV variants may coexist in a plant in the case of sequential inoculation. 1. McKinney, H.H. (1929). Mosaic diseases in the Canary Islands, West Africa and Gibraltar. Journal of Agricultural Research 39, 577–578. 2. Folimonova, S.Y., Robertson, C.J., Shilts, T., Folimonov, A.S., Hilf, M.E., Garnsey, S.M., and Dawson, W.O. (2010). Infection with strains of Citrus tristeza virus does not exclude superinfection by other strains of the virus. JVI 84, 1314–1325. https://doi.org/10.1128/JVI.02075-09. 3. Agüero, J., Gómez-Aix, C., Sempere, R.N., García-Villalba, J., García-Núñez, J., Hernando, Y., and Aranda, M.A. (2018). Stable and broad spectrum cross-protection against pepino mosaic virus attained by mixed infection. Front. Plant Sci. 9, 1810. https://doi.org/10.3389/fpls.2018.01810. 4. Harper, S.J. (2013). Citrus tristeza virus: evolution of complex and varied genotypic groups. Front. Microbiol. 4. https://doi.org/10.3389/fmicb.2013.00093.