Fig. 5.

Loss of CpG dinucleotides and ZAP-binding motifs in SARS-CoV-2 genomes: (a) Graphs showing the location and conservation of ZAP-binding motifs in the SARS-CoV-2 genome. Genes in the SARS-CoV-2 genome are indicated in different colors. (b) All ZAP-binding motifs end with a terminal CpG dinucleotide. The percentage of sequences that lost ZAP-binding motifs and the terminal CpG within the motifs are plotted. The loss of ZAP-binding motifs from the SARS-CoV-2 genomes is primarily associated with the loss of the terminal CpG within motifs. (c) The number of sequences that lost CpG dinucleotides in the SARS-CoV-2 genome were analyzed. The extent of CpG depletion from the terminal CpG site within the ZAP-binding motif is comparable to that from the other CpG sites (i.e., all CpGs other than the terminal CpG site within ZAP-binding motifs) in the SARS-CoV-2 genome. Since ZAP-binding motifs may be overlapping, only nonoverlapping or unique CpG sites are considered. (d) The loss of CpGs from all the CpG sites in ZAP-binding motifs (includes the terminal CpG and all other CpGs within ZAP-binding motifs) is comparable to that from all the CpG sites outside ZAP-binding motifs in the SARS-CoV-2 genomes. Since ZAP-binding motifs may be overlapping, only nonoverlapping or unique CpG sites are considered for the violin plots. The Mann–Whitney U test was used to compare the medians in the violin plots. ZAP, zinc-finger antiviral protein.