Figure 6
Repeat type as a function of MS diversity. For all MS markers combined, the box plots compare the mean number (no) of alleles per locus (y-axis, blue), expected heterozygosity (He) (y-axis, red) and maximum repeat length (y-axis, green) across different repeat types (x1°-axis), stratified API (x2°-axis). Character values (A and B) denote statistical significance between perfect and non-perfect repeat types, within each API category. In the API >0.05 stratum, non-perfect repeat types had a significantly higher mean no of alleles per locus (panel i) (p <0.0001, ANOVA), mean He(panel iii) (p = 0.0192, ANOVA) and mean maximum repeat length (panel v) (p <0.0001, ANOVA) than perfect repeat types. In the API ≤0.05 stratum, non-perfect repeat types had a significantly higher mean maximum repeat length (panel vi) (p <0.0001, ANOVA) than perfect repeat types. No significant differences were found between repeat types for either the mean no of alleles per locus (panel ii) or mean He in the API ≤0.05 stratum (panel iv).