Summary of selective sweep analyses
Common selective sweeps were detected among both Kauai and Bermuda feral
populations, at least two of which involved loci and variants that were
not found in earlier studies of domestication or breed improvement
selection. Haplotypes in this narrow set of shared (Bermuda and Kauai)
feral sweeps chiefly exhibit signs of artificially selected (vs. Red
Junglefowl) origin, though Red Junglefowl donorship is also evident in
at least one case. However, in the case of Bermuda, the comparison with
the most shared regions appeared to consistently be the Hawaiian
population sample. Once again, this highlights the genetic similarity
between these two distinct feral populations, and a feasible role for
shared recent selection in the filtering and/or allele-frequency shifts
of contemporary feral gene pools.
When the panel of candidate sweeps from Bermuda chickens is considered
alone, very little overlap is found with recent improvement/
domestication-related selective sweeps, a pattern previously reported in
Kauai ferals. Thus, the available data suggest feral populations are
more likely to share selective sweep loci with each other than with
contemporary domesticated breeds, though selected polymorphisms may
still share the same biological pathways, even though they occur in
different genes. This confirms that feralisation is not a mere reversal
of artificially-selected genomic changes, but also highlights the
limitations of genomic analyses that lack representation from
historically present breeds and landraces – which could further
elucidate apparent connectivity between allopatric feral populations.
Additionally, the majority of Bermudian selective sweeps involved
different genomic loci to those found in the Kauai gene pool. Thus,
Bermuda and Kauai feral chickens’ local environments, ancestries, and/or
stochastic processes have taken the two populations down somewhat
divergent evolutionary paths, though selected polymorphisms may still
share the same biological pathways, even though they occur in different
genes. Nonetheless, further study may reveal functional overlap in
selected polymorphisms of these, and other, feral populations (e.g. via
changes at divergent loci controlling shared biological pathway/s and
traits).
Despite the overarching differences in feral sweep loci between Kauai
and Bermuda, the significant overlap of a subset of selective sweeps
between these two populations offers intriguing evidence of potential
evolutionary parallelism. Future genotype-phenotype studies, both for
traits involving shared feral sweeps and those produced by potentially
convergent ones, can help determine if feral genomics can yield
predictable outcomes of feralisation-related selection.