Comparative genomic analysis reveals new evidence of genus boundary for family Iridoviridae and explores qualified hallmark genes

Abstract

Members of the family Iridoviridae (iridovirids) are globally distributed and trigger adverse economic and ecological impacts on aquaculture and wildlife. Iridovirids taxonomy has previously been studied based on a limited number of genomes, but this is not suitable for the current and future virological studies as more iridovirids are emerging. In our study, 57 representative iridovirids genomes were selected from a total of 179 whole genomes available on NCBI. Then 18 core genes were screened out for members of the family Iridoviridae. Average amino acid sequence identity (AAI) analysis indicated that a cut-off value of 70% is more suitable for the current iridovirids genome database than ICTV-defined 50% threshold to better clarify viral genus boundaries. In addition, more subgroups were divided at genus level with the AAI threshold of 70%. This observation was further confirmed by genomic synteny analysis, codon usage preference analysis, genome GC content and length analysis, and phylogenic analysis. According to the pairwise comparison analysis of core genes, 9 hallmark genes were screened out to conduct preliminary identification and investigation at the genus level of iridovirids in a more convenient and economical manner.

Keywords: Codon usage; Core genes; Iridoviridae; Phylogenetics; Synteny analysis; Taxonomy.

Graphical abstract

Fig. 1

(A) The AAI network built using the genomes of 196 Iridoviridae viruses. The edge represents AAI ≥ 99% between two nodes, and each node and color represent one genome and a cluster, respectively. (B) The viral proteomic tree (ViPTree) based on whole genome sequences. Different colored branches and outermost circles indicate different clusters. Branch length indicates evolutionary distance.

Fig. 2

The BLASTp network of top 28 orthogroups. Each node represents one amino acid sequence. The edge represents percentage of identical matches >0 between two nodes (E-value threshold of 1e-5). Core genes defined by Eaton are colored.

Fig. 3

(A) AAI network of 57 iridovirids genomes (A1: cut-off ≥ 50%, A2: cut-off ≥ 70%). Each node represents one genome. Nodes connected by lines indicate that the AAI value of connected nodes is ≥ 50% or 70%. (B) Violin plot of overall identity analysis of 6922 CDS encoded by 57 iridovirids genomes. Each point represents an identity value.

Fig. 4

(A) Violin plots of the percentage of identical matches of amino acid sequences and nucleic acid sequences of core genes. Each point represents the percentage of identical matches between the two aligned sequences (Left). Points have been removed for clarity of observation (Right). (B) Synteny analysis of representative iridovirids amino acid sequences (identity threshold at 75%). Each corresponding block represents the collinear comparison of two viruses. If there were no collinear amino acid sequence at a 75% identity between two viruses, the block would be blank.

Figure 5

Phylogenic tree of iridovirids. Maximum likelihood analysis based on concatenated core genes of representative iridovirids (best-fit model according to BIC: Q.yeast + R6). The tree was rooted on midpoint. The first column of colored branches and bars represents ICTV classified iridovirids genera. The second column of colored bars represents genera or subgroup classified in this study (AAI identity cut-off ≥ 75%). The third column of colored bars represents a heat map of the GC content of the viral genome (Fig. S1). The grey bars in the last column represent viral genome size (Fig. S1). Branch length indicates evolutionary distance. The size of the point on the branch represents the bootstrap value >75.

Fig. 6

Correspondence analysis of Ranavirus, Megalocytivirus, Chloriridovirus, Lymphocystivirus, and Iridovirus. Each dot represents the RSCU value of one gene. Density statistics for the two axes are shown above and to the right of the plot, respectively.

Fig. 7

The relationship between the ENC values and GC3s. Each dot represents the ENC value (Y axes) and GC3 value (X axes) of one gene. The solid line indicates the expected curve of ENC and GC3 only in the absence of natural selection. Points on or close to the expected curve mean that the bias is caused by mutation pressure, while points below the curve indicate the presence of other influential factors such as natural selection. Density statistics for the two axes are shown above and to the right of the plot, respectively.

Supplementary figure 1