Nucleotide diversity of synonymous and non-synonymous variants in Arabidopsis lyrata Plech population

To investigate patterns of nucleotide diversity, we calculated the ratio of nonsynonymous to synonymous diversity ( πN/πS ). First, we analyzed the genetic variation in 17 individuals from the Plech population of Arabidopsis lyrata. I mapped the sequencing reads to the NT1 reference genome from Kolesnikova et al. (2023) using Bowtie2 and performed variant calling using GATK v4.6.1 (McKenna et al, 2010). This dataset serves as the foundation for downstream analyses, such as site frequency spectrum (SFS) estimation and demographic inference. The unfolded SFS was generated using easySFS, the output was directly compatible with demographic inference tools such as dadi. 
For this analysis, I used the variant calling result containing both variant and non-variant sites to compute nucleotide diversity (π) for all sites using pixy, a tool specifically designed to account for invariant sites in diversity calculations. Next, we annotated all the positions for 0-fold (nonsynonymous) and 4-fold (synonymous) sites using degenotate. Subsequently, I calculated mean  πN and mean  πS  for each gene, creating a final table of genes with their respective  πN  and  πS  values. These genes were further categorized into groups based on genetic variance components, such as additive genetic variance and  dominance variance. For each group, I performed 1000 bootstrap resampling iterations, calculating the mean  πN/πS ratio for each bootstrap. This enabled the generation of box plots visualizing the distribution of bootstrapped  πN/πS  ratios across groups, providing insights into patterns of selective pressure and molecular evolution.