4.2 Genetic diversity within demes
We find a strikingly lower level of genetic variation within Deme II than Deme I for all variability measures (Tables 1, 2), and genomic data from individual sequencing estimated much higher inbreeding levels in Deme II than in Deme I (Figure 6a, b). Similar degrees of difference in diversity appear rarely observed in sympatric salmonids (Jorde et al., 2018).
Relatively few studies estimate inbreeding from whole-genome sequencing data in natural populations. Those that do typically focus on extinct or highly threatened populations such as woolly mammoth (Palkopoulou et al., 2015), Scandinavian wolves (Kardos et al., 2018), and gorillas (van der Valk, 2019). In these populations considerably longer (>2 Mb) runs of homozygosity than ours were observed. We find only a few ROHs above 2 Mb in our populations. Deme II individuals have the highest number with a total of 14 ROHs > 2Mb (Table S7). This might imply that recent inbreeding is not pronounced in any of the populations studied here and that inbreeding is mainly due to deep historical shared ancestors of parents. Similar observations with few long ROHs suggesting limited recent inbreeding have been observed in wild Ficedulaflycatcher populations (Kardos et al., 2017).
Our estimates of F ROH range from 0.032 in individual 1 in Deme I in Lakes Bunnersjöarna to 0.405 in individual 1 in Deme II (Figure 6b). The samples from other lakes show values between these extremes but all values for other lakes are aboveF ROH>0.09 with an average of 0.18 (Table S7). This appears high, and is above what has been estimated for hatchery strains of rainbow trout where the highest observations were typically around 0.1-0.2 (D’Ambrosio et al., 2019). A large extent of this difference might be due to differences in settings of ROH analyses: D´Ambrosia et al. (2019) allowed a maximum of one heterozygous genotype per ROH while we allowed three. For a low coverage data like ours, a maximum of three heterozygous genotype is recommended (e.g. Ceballos et al., 2018).