Genome-Wide Diversity Analysis of African Swine Fever Virus Based on a Curated Dataset

Abstract

African swine fever (ASF) is a lethal contagious viral disease of domestic pigs and wild boars caused by the African swine fever virus (ASFV). The pandemic spread of ASF has had serious effects on the global pig industry. Virus genome sequencing and comparison play an important role in tracking the outbreaks of the disease and tracing the transmission of the virus. Although more than 140 ASFV genome sequences have been deposited in the public databases, the genome-wide diversity of ASFV remains unclear. Here we prepared a curated dataset of ASFV genome sequences by filtering genomes with sequencing errors as well as duplicated genomes. A total of 123 ASFV genome sequences were included in the dataset, representing 10 genotypes collected between 1949 and 2020. Phylogenetic analysis based on whole-genome sequences provided high-resolution topology in differentiating closely related ASFV isolates, and drew new clues in the classification of some ASFV isolates. Genome-wide diversity of ASFV genomes was explored by pairwise sequence similarity comparison and ORF distribution comparison. Tandem repeat sequences were found widely distributed and highly varied in ASFV genomes. Structural variation and highly variable poly G or poly C tracts also contributed to the genome diversity. This study expanded our knowledge on the patterns of genetic diversity and evolution of ASFV, and provided valuable information for diagnosis improvement and vaccine development.

Keywords: African swine fever virus; curated dataset; genome sequence diversity; tandem repeat sequences.

Figure 1

Distribution of African swine fever virus genome sequences included in the curated dataset. (A) Map indicates the collection locations of the ASFV genome sequences. (B) Distribution of the host of the ASFV genome sequences. (C) Distribution of the collection date of the ASFV genome sequences.

Figure 2

Maximum likelihood phylogenetic tree of ASFV genome sequences in the curated dataset. The tree is midpoint rooted. The scale bar is given in numbers of substitutions per site. Bootstrap resampling (1000 replications) support values are shown at the nodes.

Figure 3

Sequence similarity matrix plot of ASFV genome sequences in the curated dataset. The level of identity of pairwise genome sequences is indicated by different colors. Dark red represents 100% identity, and blue represents lower identity. The maximum likelihood phylogenetic tree of ASFV genome sequences is shown on the right.

Figure 4

Structural variations in ASFV genotype I (A) and genotype II (B) genomes. Fragment insertion is shown in dark grey blocks. Fragment deletion is shown in dashed blank blocks. Blank blocks indicate gaps introduced for alignment purposes. Nucleotide position in the reference genome L60 (A) and genome Georgia 2007/1 (B) is shown.

Figure 5

Sequence variation of the tandem repeat sequences (TRS) in ASFV genomes. (A) Variation of TRS in other regions in ASFV genotype I genomes. (B) Variation of TRS in the coding region of B602L gene in ASFV genotype I genomes. (C) Variation of TRS in ASFV genotype II genomes. The arrangement of the TRS in each genome was listed according to the nucleotide sequence. Each arrow of a different color represents a type of repeat unit with a specific nucleotide sequence. The nucleotide sequence of each type of repeat unit was also shown. The amino acid sequence of each type of repeat unit in the TRS in the coding region of B602L gene was also shown.

Figure 6

Sequence variation of the highly variable poly G or poly C repeats sequence in ASFV genotype I (A) and genotype II (B) genomes. Nucleotide position in the reference genome L60 (A) and genome Georgia 2007/1 (B) is shown.