Skip to main content
. 2017 Nov 14;8:1958. doi: 10.3389/fpls.2017.01958

FIGURE 4.

FIGURE 4

Bivariate relationships between MeRV prevalence/population evolutionary parameters and the best predictor variable. In each case, best predictor variable was identified by model selection analyses. Regressions of host density (A) and rainfall (B) on MeRV prevalence; and regressions of MeRV population haplotype diversity (C), population genetic diversity (D), number of synonymous mutations per site (E) and overall selection pressures dN/dS (F) on MeRV prevalence (% of mountain rue infected plants) are represented. Note the different scales on the axes depending on the specific parameter analyzed.