The effect of population size on short-term rates of evolution in natural populations

Population size is a key factor for many fundamental ecological, genetic and evolutionary processes. An important question is if and how fast small populations will be able to evolve in response to changes in the environment. In this project we aimed at integrating ecological and genetic processes to examine how population size may affect short-term rates of adaptive evolution in natural populations. Unique empirical data from long-term studies of natural and experimental house sparrow (Passer domesticus) populations at the coast of Helgeland have been collected and analyzed with state-of-the art genomics tools and advanced statistical methods to try to answer these questions. Our results show that small populations lost genetic variation faster than large populations, and that genetic variation was lost faster than expected from the observed population size. Variation in reproductive success among individuals, and especially among older males, seemed to be the most important factor affecting how much faster genetic variation was lost. Immigration did however appear to be able to counteract the loss of genetic variation, and dispersal also acted as a strong force to make populations more genetically similar. An interaction between dispersal and fluctuations in population size was probably the reason why we found that populations on islands tended to have lower levels of genetic variation and were more differentiated than populations on the mainland. We used results from a translocation experiment at the Helgeland coast where house sparrows were moved from Vega to Vikna to examine how "alien" genes spread in a population. The analyses showed that the genetic footprint of introduced individuals was relatively small, perhaps due to social processes within each sub-population, but that genetic variation introduced from Vega was still present after almost a decade. We also found that the genetic composition of the Vikna population varied over time, but that this probably was due to random processes within the relatively small sub-populations. During the project we have developed advanced statistical models to estimate the relative importance of genes and environment for variation in different types of traits. Using such models we have shown that the heritable component of some traits, and also that the genetic link between traits may differ between populations. These results suggest that the rate of adaptive evolutionary change may differ between populations of the same species, even when selection is acting in the same way. Our statistical genetic models can also be used to examine if any observed changes in a trait over time is due to changes in genes coding for the trait or a result of changes in the environment. For example, we could determine that the reduction in bill depth which we observed from 1993 to 2002 in some of our study populations was likely due to environmental and not genetic changes. Selection, a relationship between phenotype and fitness (survival and/or reproduction), is the main force that causes adaptive evolution of heritable traits. Our studies have demonstrated that for example parasite load and metabolic rate affect individual fitness, but that their effects on fitness may differ between populations and sexes. Furthermore, we have used data from an artificial selection experiment to show that there seems to be an optimum body size in house sparrows at Helgeland, and that natural processes force phenotypes to remain close to this optimum. We have also developed advanced statistical methods that give correct estimates of selection in natural populations, which usually have individuals of different age classes and that are subject to random fluctuations in both individual fitness (may be large in smaller populations) and the environment. These models make the appropriate links between ecological and evolutionary processes in natural populations. To further understand the genetic and evolutionary processes in our model system we have generated a house sparrow reference genome sequence, and genotyped 200,000 genetic markers called Single Nucleotide Polymorphisms (SNPs) spread across the genome in ca. 4000 individuals. This work was very successful, and our high-density genotype data has been used to examine the genetic architecture of several morphological traits. These analyses have for example shown that the proportion of phenotypic variance explained by different chromosomes is related to the size of each chromosome, suggesting that most traits are affected by many genes spread across the genome. However, we have also tried to map the location of any genes that affect morphological traits, and were able to find particular places in the genome, called quantitative trait loci (QTL), where it is likely that genes coding for the traits are located. This may open up the possibility to study how ecological and evolutionary processes act directly on individual genes.

Population size is the key factor governing many fundamental ecological, population genetic and evolutionary processes. A major concern is to what extent small populations are able to evolve in response to changes in the environment. Our goal is to integr ate ecological and population genetic processes to examine how population size and spatio-temporal scaling of environmental fluctuations and dispersal affect short-term rates of adaptive evolution in natural age-structured populations in fluctuating envir onments. Unique empirical data from long-term studies of natural and experimental house sparrow (Passer domesticus) populations will be analysed with state-of-the art genomics tools and advanced statistical methods. The results will enable us to examine h ow key processes important to short-term adaptive evolution are affected by variation in population size and dispersal. Specifically we will: 1) Estimate population specific levels of demographic and environmental stochasticity to provide a link between p opulation ecological and population genetic processes. 2) Examine effects of population size and dispersal on actual rates of genetic drift across the genome. 3) Investigate how population size and dispersal affects levels of additive genetic variance and quantify the genetic load it may cause in a fluctuating environment. 4) Estimate strength and direction of selection across populations and years and examine how population size and dispersal affect spatio-temporal variation in selection, and use these w ith estimates of additive genetic variance to examine effects on predicted rates of adaptive evolution. 5) Test how population size and dispersal affect actual rates of adaptive evolution estimated by changes in estimated breeding values and genomic breed ing values. 6) Examine how population size and dispersal affect spatio-temporal variation in selection and contemporary rates of evolution measured at SNPs within individual quantitative trait loci.