Stefan Palm

Power for detecting genetic divergence: differences between statistical methods and marker loci

Information on statistical power is critical when planning investigations and evaluating empirical data, but actual power estimates are rarely presented in population genetic studies. We used computer simulations to assess and evaluate power when testing for genetic differentiation at multiple loci through combining test statistics or P values obtained by four different statistical approaches, viz. Pearsons chi‐square, the log‐likelihood ratio G‐test, Fishers exact test, and an FST‐based permutation test. Factors considered in the comparisons include the number of samples, their size, and the number and type of genetic marker loci. It is shown that power for detecting divergence may be substantial for frequently used sample sizes and sets of markers, also at quite low levels of differentiation. The choice of statistical method may be critical, though. For multi‐allelic loci such as microsatellites, combining exact P values using Fishers method is robust and generally provides a high resolving power. In contrast, for few‐allele loci (e.g. allozymes and single nucleotide polymorphisms) and when making pairwise sample comparisons, this approach may yield a remarkably low power. In such situations chi‐square typically represents a better alternative. The G‐test without Williamss correction frequently tends to provide an unduly high proportion of false significances, and results from this test should be interpreted with great care. Our results are not confined to population genetic analyses but applicable to contingency testing in general.

Genetic Population Structure of Fishes: Implications for Coastal Zone Management

Abstract The pattern for distribution of genetic variation within and between populations is referred to as the genetic population structure of the species. To avoid depletion of genetic resources sustainable management should be based on knowledge of this structure. We discuss key aspects of genetic population structure in the context of identifying biological units for fisheries management, suggesting three basic types of structuring: distinct populations; continuous change; and no differentiation. The type of structure determines how units for genetically sustainable management are to be identified. We also review what is currently known regarding the genetic population structure of fishes exploited in the Swedish part of the Baltic Sea, and conclude that sufficient genetic information is lacking for most of the species. This is a serious problem, particularly considering that populations of several commercially exploited fishes are declining and some exhibit recruitment problems. For six species, Atlantic herring, Atlantic salmon, brown trout, European eel, turbot, and pike, sufficient genetic data are available to provide at least basic information on genetic structure and genetic units for biologically sustainable use. Current management practices do not sufficiently consider these data.

Concordance of allozyme and microsatellite differentiation in a marine fish, but evidence of selection at a microsatellite locus

Previous studies have reported higher levels of divergence for microsatellites than for allozymes in several species, suggested to reflect stabilizing selection on the allozymes. We compared the differentiation patterns of 11 allozyme and nine microsatellite loci using 679 spawning Atlantic herring (Clupea harengus) collected in the Baltic and North Seas to test for differential natural selection on these markers. Observed distributions of F statistics for the two types of markers are conspicuously dissimilar, but we show that these differences can largely be explained by sampling phenomena caused by different allele frequency distributions and degrees of variability. The results show consistently low levels of differentiation for both marker types, with the exception of one outlier microsatellite locus with a notably high FST. The aberrant pattern at this locus is primarily due to two alleles occurring at markedly high frequencies in the Baltic, suggesting selection at this locus, or a closely linked one. When excluding this locus, the two marker types show similar, weak differentiation patterns with FST values between the Baltic and the North Seas of 0.001 and 0.002 for allozymes and microsatellites, respectively. This small heterogeneity, and weak isolation by distance, is easier to distinguish statistically with microsatellites than with allozymes that have fewer alleles and skewed frequency distributions. The allozymes, however, also detect surprisingly low levels of divergence. Our results support suggestions that previously described differences between marker types are primarily caused by a small number of outlier loci.

Spatial genetic structure of northern pike (Esox lucius) in the Baltic Sea.

The genetic relationships among 337 northern pike (Esox lucius) collected from the coastal zone of the central Baltic region and the Finnish islands of Åland were analysed using five microsatellite loci. Spatial structure was delineated using both traditional F‐statistics and individually based approaches including spatial autocorrelation analysis. Our results indicate that the observed genotypic distribution is incompatible with that of a single, panmictic population. Isolation by distance appears important for shaping the genetic structure of pike in this region resulting in a largely continuous genetic change over the study area. Spatial autocorrelation analysis (Morans I) of individual pairwise genotypic data show significant positive genetic correlation among pike collected within geographical distances of less than c. 100–150 km (genetic patch size). We suggest that the genetic patch size may be used as a preliminary basis for identifying management units for pike in the Baltic Sea.

Lack of molecular genetic divergence between sea-ranched and wild sea trout (Salmo trutta).

The supportive breeding programme for sea trout (Salmo trutta) in the River Dalälven, Sweden, is based on a sea‐ranched hatchery stock of local origin that has been kept ‘closed’ to the immigration of wild genes since the late 1960s (about seven generations). In spite of an apparent potential for substantial uni directional gene flow from sea‐ranched to wild (naturally produced) trout, phenotypic differences with a presumed genetic basis have previously been observed between the two ‘stocks’. Likewise, two previous studies of allozyme and mitochondrial DNA variation based on a single year of sampling have indicated genetic differentiation. In the present study we used microsatellite and allozyme data collected over four consecutive years, and tested for the existence of overall genetic stock divergence while accounting for temporal heterogeneity. Statistical analyses of allele frequency variation (F‐statistics) and multilocus genotypes (assignment tests) revealed that wild and sea‐ranched trout were significantly different in three of four years, whereas no overall genetic divergence could be found when temporal heterogeneity among years within stocks was accounted for. On the basis of estimates of effective population size in the two stocks, and of FST between them, we also assessed the level of gene flow from sea‐ranched to wild trout to be ≈ 80% per generation (with a lower confidence limit of ≈ 20%). The results suggest that the reproductive success of hatchery and naturally produced trout may be quite similar in the wild, and that the genetic characteristics of the wild stock are largely determined by introgressed genes from sea‐ranched fish.

Panmixia in European eel revisited: no genetic difference between maturing adults from southern and northern Europe

Previous studies of genetic structure in the European eel have resulted in seemingly conflicting results, ranging from no detectable heterogeneity to small but statistically significant differences and isolation by distance patterns among eels sampled across the continental range. Differences with respect to sampling design and choice of molecular markers, combined with a lack of power estimates, complicate comparisons of existing results. In this study we have used six microsatellite markers and, for the first time, compared maturing silver eels of known age from southern and northern Europe (Italy and Baltic Sea). In comparison with previous studies, our data may give a better representation of potential spawning stocks because eels were sampled when having begun their migration toward the presumed spawning area in the Sargasso Sea. Despite large sample sizes (404 and 806 individuals) we could not observe any signs of genetic differentiation (average FST=−0.00003, P=0.61), and a power analysis showed that the true level of heterogeneity (if existing) must be exceedingly small to have remained undetected (say, FST <0.0004). A tendency for slightly increased genetic differences between cohorts over time could be seen, but the amount of temporal change was minor and not statistically significant. Our findings reiterate the notion that previous reports of continental genetic differentiation in the European eel may be largely explained by uncontrolled temporal variation between juvenile glass eel samples.

Estimating genetic drift and effective population size from temporal shifts in dominant gene marker frequencies

Measurement of temporal change in allele frequencies represents an indirect method for estimating the genetically effective size of populations. When allele frequencies are estimated for gene markers that display dominant gene expression, such as, e.g. random amplified polymorphic DNA (RAPD) and amplified fragment length polymorphism (AFLP) markers, the estimates can be seriously biased. We quantify bias for previous allele frequency estimators and present a new expression that is generally less biased and provides a more precise assessment of temporal allele frequency change. We further develop an estimator for effective population size that is appropriate when dealing with dominant gene markers. Comparison with estimates based on codominantly expressed genes, such as allozymes or microsatellites, indicates that about twice as many loci or sampled individuals are required when using dominant markers to achieve the same precision.

Microsatellite markers reveal clear geographic structuring among threatened noble crayfish (Astacus astacus) populations in Northern and Central Europe

Noble crayfish (Astacus astacus L.), the most highly valued freshwater crayfish in Europe, is threatened due to a long-term population decline caused mainly by the spread of crayfish plague. Reintroduction of the noble crayfish into restored waters is a common practice but the geographic and genetic origin of stocking material has rarely been considered, partially because previous genetic studies have been hampered by lack of nuclear gene markers with known inheritance. This study represents the first large scale population genetic survey of the noble crayfish (633 adults from 18 locations) based on 10 newly developed microsatellite markers. We focused primarily on the Baltic Sea area (Estonia, Finland and Sweden) where the largest proportion of the remaining populations exists. To allow comparisons, samples from the Black Sea catchment (the Danube drainage) were also included. Two highly differentiated population groups were identified corresponding to the Baltic Sea and the Black Sea catchments, respectively. The Baltic Sea catchment populations had significantly lower genetic variation and private allele numbers than the Black Sea catchment populations. Within the Baltic Sea area, a clear genetic structure was revealed with population samples corresponding well to their geographic origin, suggesting little impact of long-distance translocations. The clear genetic structure strongly suggests that the choice of stocking material for re-introductions and supplemental releases needs to be based on empirical genetic knowledge.

Strong divergence in trait means but not in plasticity across hatchery and wild populations of sea‐run brown trout Salmo trutta

There is ample evidence that organisms adapt to their native environment when gene flow is restricted. However, evolution of plastic responses across discrete environments is less well examined. We studied divergence in means and plasticity across wild and hatchery populations of sea‐run brown trout (Salmo trutta) in a common garden experiment with two rearing environments (hatchery and a nearly natural experimental stream). Since natural and hatchery environments differ, this arrangement provides an experiment in contemporary adaptation across the two environments. A QST − FST approach was used to investigate local adaptation in survival and growth over the first summer. We found evidence for divergent selection in survival in 1 year and in body length in both years and rearing environments. In general, the hatchery populations had higher survival and larger body size in both environments. QST in body size did not differ between the rearing environments, and constitutive divergence in the means was in all cases stronger than divergence in the plastic responses. These results suggest that in this system, constitutive changes in mean trait values are more important for local adaptation than increased plasticity. In addition, ex situ rearing conditions induce changes in trait means that are adaptive in the hatchery, but potentially harmful in the wild, suggesting that hatchery rearing is likely to be a suboptimal management strategy for trout populations facing selection in the stream environment.

Adaptive divergence in body size overrides the effects of plasticity across natural habitats in the brown trout

The evolution of life-history traits is characterized by trade-offs between different selection pressures, as well as plasticity across environmental conditions. Yet, studies on local adaptation are often performed under artificial conditions, leaving two issues unexplored: (i) how consistent are laboratory inferred local adaptations under natural conditions and (ii) how much phenotypic variation is attributed to phenotypic plasticity and to adaptive evolution, respectively, across environmental conditions? We reared fish from six locally adapted (domesticated and wild) populations of anadromous brown trout (Salmo trutta) in one semi-natural and three natural streams and recorded a key life-history trait (body size at the end of first growth season). We found that population-specific reaction norms were close to parallel across different streams and QST was similar – and larger than FST – within all streams, indicating a consistency of local adaptation in body size across natural environments. The amount of variation explained by population origin exceeded the variation across stream environments, indicating that genetic effects derived from adaptive processes have a stronger effect on phenotypic variation than plasticity induced by environmental conditions. These results suggest that plasticity does not “swamp” the phenotypic variation, and that selection may thus be efficient in generating genetic change.