Variant-specific mutational spectra

[szhan]

We examined the mutational spectrum of the entire SARS-CoV-2 ARG in our preprint, but the mutational spectrum has been shown to differ by variant of concern.

Bloom et al. (2023)) has shown that the mutational spectra at 4-fold degenerate sites are different in Alpha, Delta, and Omicron BA.1. The choice of colours in Figure 1 makes it hard to discern the differences across all the mutation types, but the frequency difference for G>T stands out - that is, G>T occurs more frequently in Alpha and Delta (> 0.10) than Omicron BA.1 (< 0.10) (Figure 1B, left), again at 4-fold degenerate sites.

Just out of curiosity, I took a look the variant-specific mutational spectra in the temporary ARG viridian-full-mm_3-mhc_5-mgs_20-2021-03-21.ts.tsz. Even though I did not focus on 4-fold degenerate sites, I observed that the frequency of G>T is higher in Alpha (0.12) and Delta (0.20) than Omicron BA.1 (0.04).

In our mutational spectra, there is a pretty noticeable difference in the frequency of 'C>T' - it occurs more frequently in Alpha (0.52) and Delta (0.40) than Omicron BA.1 (0.33). There doesn't seem to be a such a noticeable difference in the Bloom et al. mutational spectra, though it may be because I am not focusing on the 4-fold degenerate sites.

[szhan]

Tagged this file here because it indicates which sites are 4-fold degenerate and which were excluded in the analysis.

https://media.githubusercontent.com/media/jbloomlab/SARS2-mut-spectrum/refs/heads/main/results/mutation_counts/aggregated.csv