Extended Data Fig. 9: Signatures of selection in the PLA2 venom gene region.
From: The roles of balancing selection and recombination in the evolution of rattlesnake venom
Signatures of selection in the PLA2 venom gene region, with comparisons to chromosomal backgrounds and non-venom homologues (NV). a Tajima’s D across chromosome 15 in C. viridis (CV1) and C. oreganus (CO1) populations (top). Middle panels show variation zoomed into the PLA2 region, shaded in grey. Boxplots show distributions of Tajima’s D for chromosome 15, non-venom homologues of the PLA2 family, and PLA2s. b Proportion of fixed differences (df) between CV1 and CO1. c Integrated haplotype statistics (|iHS|) in CV1 and CO1. d ß statistic measuring allele frequency correlation in CV1 and CO1. Points in chromosome scan panels represent mean estimates in 10 kb sliding windows, and lines represent 100 kb windowed estimates. Points in zoomed PLA2 region scans represent mean estimates in 250 bp windows and lines show 1 kb windowed estimates. Locations of individual PLA2 genes are shown as dark blue segments (bottom panels). The PLA2gIIE non-venom homologue is shown in light purple. Gaps in lines represent windows where there was insufficient data to calculate a mean estimate. Boxplots in a-d show the median (horizontal lines), interquartile (box limits), and range (whiskers). Asterisks indicate significant differences between venom gene regions and chromosome backgrounds and non-venom homologues based on two-tailed Welch’s two-sample t-tests (Tajima’s D and |iHS|) and Mann–Whitney ∪ tests (df and ß) and n = 1,272 10 kb sliding windows for comparisons (*P < 0.05; **P < 0.001; ***P < 2.2 × 10−16). Exact P-values for comparisons can be found in Supplementary Table 5. e-f Results of composite-likelihood ratio tests of balancing selection versus neutrality using B0,MAF scores in the PLA2 region in CV1 and CO1. Upper panels show B0,MAF scores, with higher values indicating greater evidence for balancing selection. Arrows show locations of PLA2 genes. Lower panels show inferred footprint size, log\(\widehat {(A)}\), as solid grey lines and equilibrium allele frequency, \(\hat x\), as dashed lines. Dashed lines in top panels of e-f show the genome-wide 95th quantile.
