JSON
Arnqvist G, Sayadi A
Ecol Evol 12 (10) e9440 [2022-10-00; online 2022-10-27]
Efforts to unravel the genomic basis of incipient speciation are hampered by a mismatch between our toolkit and our understanding of the ecology and genetics of adaptation. While the former is focused on detecting selective sweeps involving few independently acting or linked speciation genes, the latter states that divergence typically occurs in polygenic traits under stabilizing selection. Here, we ask whether a role of stabilizing selection on polygenic traits in population divergence may be unveiled by using a phenotypically informed integrative approach, based on genome-wide variation segregating in divergent populations. We compare three divergent populations of seed beetles (Callosobruchus maculatus) where previous work has demonstrated a prominent role for stabilizing selection on, and population divergence in, key life history traits that reflect rate-dependent metabolic processes. We derive and assess predictions regarding the expected pattern of covariation between genetic variation segregating within populations and genetic differentiation between populations. Population differentiation was considerable (mean F ST = 0.23-0.26) and was primarily built by genes showing high selective constraints and an imbalance in inferred selection in different populations (positive Tajima's D NS in one and negative in one), and this set of genes was enriched with genes with a metabolic function. Repeatability of relative population differentiation was low at the level of individual genes but higher at the level of broad functional classes, again spotlighting metabolic genes. Absolute differentiation (d XY) showed a very different general pattern at this scale of divergence, more consistent with an important role for genetic drift. Although our exploration is consistent with stabilizing selection on polygenic metabolic phenotypes as an important engine of genome-wide relative population divergence and incipient speciation in our study system, we note that it is exceedingly difficult to firmly exclude other scenarios.
NGI Uppsala (SNP&SEQ Technology Platform) [Service]
National Genomics Infrastructure [Service]
PubMed 36311399
DOI 10.1002/ece3.9440
Crossref 10.1002/ece3.9440
pmc: PMC9608792
pii: ECE39440