The low abundance of CpG in the SARS-CoV-2 genome is not an evolutionarily signature of ZAP

Figures

• 

  Figure 1. PCA of viral motifs representations. D-values of all dinucleotide, trinucleotide and tetranucleotide motifs in all viral sequences form a matrix, which is used as an input for PCA. (A) PC1-PC2 plot shows four clusters, one for each virus family: H1N1, coronaviruses (CoV), HBV, and HIV-1. (B) PC1-PC2 plot classifies coronaviruses into two clusters. All 664 SARS-CoV-2 (including reference sequence), Bat-RaTG13, RmYN02, 4 Bat-Coronaviridae viruses (MW703458, MW251308, MG772933, MG772934), and 10 Pangolin-Coronaviridae (EPI_ISL_410721, EPI_ISL_410539, EPI_ISL_410542, EPI_ISL_410543, EPI_ISL_410538, EPI_ISL_410541, EPI_ISL_410540, MT040336.1, MT040333.1, MT121216.1) formed a cluster (SARS-CoV-2-like group), which is separated from the rest of coronavirus sequences including Human coronavirus 229E, Bat-Coronaviridae, Human coronavirus HKU1, Murine coronavirus, MERS coronavirus, Human coronavirus NL63, Human coronavirus OC43, Primates-Coronaviridae, SARS coronavirus Tor2, SARS coronavirus Ubani, Viverrids-Coronaviridae, SARS coronavirus wtic_MB, SARS coronavirus GZ02. SARS-CoV-2-like are highlighted with a square.

• 

  Figure 2. Comparison of dinucleotide motif representations between SRAS-CoV-2-like and SARS-CoV groups. D-values of each dinucleotide were compared between the two viral groups SARS-CoV-2-like and SARS-CoV. Mann–Whitney test was used to examine the difference in the median of D-values between the two coronavirus groups. D-value (motif representation) is defined as the ratio of the observed frequency (Pobs) of a motif over its expected frequency (Pexp). Pobs is simply the observed relative frequency of the motif. Pexp is quantified using the frequency of the motif in the sequence and the frequencies of the smaller constituting motifs.

• 

  Figure 3. Co-location of ZAP binding regions and CpG motifs. Overlaying of the ZAP binding peaks and CpG densities in (A) JEV genome (Japanese Encephalitis Virus) and (B) HIV-1. The ZAP binding peaks (density of reads aligned to the genome) are estimated using a 250 bp sliding window moving by 1 bp along the viral genomes. The CpG density was calculated using the same sliding window analysis method, except we used a 200 bp window sliding by 1 bp in JEV and a 250 bp window sliding by 1pb in HIV-1. ZAP binding peaks and CpG densities are shown in green and red, respectively. # Location of CpGs. * Number of CpGs per 1 Kb.

• 

  Figure 4. Comparison of the abundance of ZAP optimal binding motif C(n7)G(n)CG with the control motif C(n7)C(n)CG in viruses of SARS-CoV-2-like group. The abundance of ZAP optimal binding motif C(n7)G(n)CG was compared to C(n7)C(n)CG in the SARS-CoV-2-like group. The motif C(n7)C(n)CG was used here as a control. Mann–Whitney test was used to determine the difference in the median of abundance between these two motifs.