Within-generation consequences of postsettlement mortality for trait composition in wild populations: An experimental test

Genomic basis for early-life mortality in sharpsnout seabream

Mortality at early life stages of fishes is common in nature and can be shaped by stochastic and selective processes. Selective mortality has rarely been assessed in natural conditions but can now be studied by combining genomic data with information on different life stages that realates to fitness. Here we investigate selective mortality between settlers and six-month survivors of the sharpsnout seabream by genotype-phenotype/environmental association studies in three localities along a geographic gradient. We gathered information on 105 individuals at 85,031 SNPs, obtained from individual based 2b-RAD libraries, as well as 9 phenotypic and environmental variables derived from individual otolith readings. We found common signals across localities for potential selection, such as lower survival rates for individuals hatching earlier, growing faster and experiencing higher temperatures during their planktonic phase. We identified 122 loci with parallel significant association to phenotypic and environmental variables. Importantly, one of these loci mapped to the exonic region of the il20rb, a gene involved in immune response, in the phylogenetically closest reference genome, showing parallel frequency changes in non-synonymous mutations in the three studied populations. Further temporal assessments are needed to understand how polymorphisms that are key to selective mortality are maintained.

Genetic and phenotypic variation exhibit both predictable and stochastic patterns across an intertidal fish metapopulation

Interactions among selection, gene flow, and drift affect the trajectory of adaptive evolution. In natural populations, the direction and magnitude of these processes can be variable across different spatial, temporal, or ontogenetic scales. Consequently, variability in evolutionary processes affects the predictability or stochasticity of microevolutionary outcomes. We studied an intertidal fish, Bathygobius cocosensis (Bleeker, 1854), to understand how space, time, and life stage structure genetic and phenotypic variation in a species with potentially extensive dispersal and a complex life cycle (larval dispersal preceding benthic recruitment). We sampled juvenile and adult life stages, at three sites, over three years. Genome-wide SNPs uncovered a pattern of chaotic genetic patchiness, that is, weak-but-significant patchy spatial genetic structure that was variable through time and between life stages. Outlier locus analyses suggested that targets of spatially divergent selection were mostly temporally variable, though a significant number of spatial outlier loci were shared between life stages. Head shape, a putatively ecologically responsive (adaptive) phenotype in B. cocosensis also exhibited high temporal variability within sites. However, consistent spatial relationships between sites indicated that environmental similarities among sites may generate predictable phenotype distributions across space. Our study highlights the complex microevolutionary dynamics of marine systems, where consideration of multiple ecological dimensions can reveal both predictable and stochastic patterns in the distributions of genetic and phenotypic variation. Such considerations probably apply to species that possess short, complex life cycles, have large dispersal potential and fecundities, and that inhabit heterogeneous environments.