Phylogenetic analysis of a Bean yellow mosaic virus isolate from Iran and selecting the phylogenetic marker by comparing the individual genes and complete genome trees of BYMV isolates

Abstract

Bean yellow mosaic virus (BYMV) is an economic potyvirus with worldwide distribution and a broad host range. Recently, phylogenetic studies of BYMV isolates have indicated the importance of comparing the full genome sequences to obtain a reliable classification. However, most available isolates in public databases are partially sequenced, and virologists continue to prefer partial genomes or individual genes for phylogenetic studies. As such, to avoid unbiased classification, phylogenetic studies should be based on either phylogenetic gene markers or a multigene dataset instead of just one random gene. In the present study, we sequenced the full genome of a BYMV isolate collected from an Iranian faba bean field using the Illumina Novaseq 6000 platform. The phylogenetic analysis revealed that the Iranian isolate is closely related to three isolates originating from faba bean fields in Japan, Australia and Germany. To identify the phylogenetic markers of BYMV, the topology distance between reconstructed trees based on individual genes and the whole genome was calculated using topology congruence tests considering only the topology of the compared trees and a character congruence test taking into account the characters, tree topology and estimated bootstrap support value. Accordingly, nuclear inclusion proteins a and b (NIa and NIb) and viral protein genome linked (VPg) were selected as evolutionary markers which can be used as appropriate representations of the full genome in phylogenetic studies. Furthermore, supermatrix approach was used to compare all BYMV isolates that have been fully or partially sequenced. The results demonstrate that all submitted BYMV strains can be divided into 10 main monophyletic clades with high bootstrap values; these are named BYM-1 through BYM-10 to avoid confusion with previously introduced nomenclature. Moreover, the results reveal that clustering based on phylogenetic analysis and their original host is not completely consistent, which supports recent phylogenetic studies based on available whole genome sequences.