2.5.2 Genetic diversity, inbreeding and mean kinship
We used the genetic_diversity function within the “gstudio” package v1.5.3 (Dyer 2021) to calculate mean (± standard error and standard deviation) genetic diversity metrics across the 10 subpopulations. These metrics included the number of alleles (A), effective number of alleles (AE), observed heterozygosity (HO), expected heterozygosity (HE). We used the standard correction for small sample size to account for potential biases in the estimates of HE due to variation in the number of samples obtained from subpopulations. We used “diveRsity” v1.9.90 (Keenan et al. 2013) to calculate subpopulation-level Wright’s inbreeding coefficients (FIS) values with confidence intervals via 1000 bootstraps. We then used “PopGenReport” v3.0.7 (Gruber & Adamack 2022) to calculate mean allelic richness (AR) in the subpopulations and “poppr” to calculate (i) mean private alleles within all subpopulations bootstrapped to sample size of 5; and (ii) the number of private alleles only within subpopulations of the King Island scrubtit. Finally, we calculated individual inbreeding coefficients (IIC) using the ‘mom.weir’ method with thesnpgdsIndInbCoef function in “SNPRelate” (Zheng et al. 2012). This method uses a modified Visscher’s estimator (Yang et al. 2010).
We calculated relatedness using COANCESTRY (Wang, 2011), firstly only within the King Island subpopulations, and then within all 10 subpopulations together (King Island and mainland Tasmania). We first ran simulations as per Hogg et al (2019) to determine the best moment estimator to use. This was the triadic likelihood method (TrioML), as it weights loci relative to the number of alleles and accounts for genotyping error and inbreeding (Wang, 2007). Results are presented as mean kinship values (MK), calculated by dividing the TrioML value by 2.