4.5 What additional information did genomic analysis provide?
The present results are consistent with patterns detected using only a few allozymes over 40 years ago. However, several new insights have been gained from the whole-genome analyses applied here. First, the results suggest that the divergence between the two demes results primarily from genetic drift, but that selection might also play a role, particularly including polygenic traits of metabolic and growth processes. Second, the pattern of divergence and the allele occurrence over the SNP array indicate that Deme II is reproductively isolated, whereas Deme I appears to be in at least some contact with Lake Ånnsjön, possibly by downstream migration. Third, the level of genome-wide diversity and inbreeding is strikingly different between the two demes and appears large also in relation to brown trout of other lakes. Deme I shows the highest degree of nucleotide diversity and lowest measures of runs of homozygosity compared to all other samples studied here, whereas Deme II shows radically lower levels of nucleotide diversity than any of the other samples, and the highest measures of ROH (cf. Table 2, Figure 6 vs. Tables S7, S12, Figures S12, S13). Further, we find that the genomic background behind the LDH-1 allozyme expressions is complex and likely involves regulatory mechanisms and possibly interactions between several genes.
This study has not addressed what causes the reproductive isolation between these demes and if this structuring has evolved sympatrically or allopatrically, if it reflects natural evolution or if human induced release has played a role. Preliminary analysis of the mitochondrial DNA sequences of these demes do not lend support for separate lineages reflecting colonization from different glaciation refugia (unpublished data). Further, we cannot exclude human translocation. For example, such translocation could imply that Deme I was translocated from one place (e.g. Lake Ånnsjön) and Deme II from another lake, and that the populations did not, or only to a very limited degree, hybridize in the new environment. There are lakes c 20 km away where the LDH-1 null allele has been observed (Allendorf et al., 1984). Further, we have not been able to address the potential temporal stability of these structures which appears highly warranted, but difficult in light of a current strict protection of these lakes.