Help Guide Disclaimer Contact us Login
  Advanced SearchBrowse





Tracing signatures of positive selection in natural populations of the house mouse


Büntge,  Anna
Department Evolutionary Genetics, Max Planck Institute for Evolutionary Biology, Max Planck Society;

Tautz,  Diethard
Department Evolutionary Genetics, Max Planck Institute for Evolutionary Biology, Max Planck Society;

There are no locators available
Fulltext (public)

(Publisher version), 9MB

Supplementary Material (public)
There is no public supplementary material available

Büntge, A. (2010). Tracing signatures of positive selection in natural populations of the house mouse. PhD Thesis, Christian-Albrechts-Universität, Kiel.

Cite as:
Understanding the genetic basis of positive selection in natural populations is one of the primary goals in evolutionary biology. Central to this aim is the identification of genomic regions that have been affected by natural selection. Two different approaches allow the investigation of traces in the genome left by selection, which have been both used to look for positive selection in natural populations of the house mouse (Mus musculus). One way is to assess candidate genes selected a priori. Such a candidate gene approach defines genes of interest based on a given phenotype, i.e. the genes are chosen on the basis of function in biochemical pathways that are relevant to specific phenotypes. This ‘top-down’ approach is advantageous if a well-defined association persists between the trait of interest and the underlying gene. In the first part of this thesis a candidate gene approach was used to study selection on detoxification genes. The analysis was based on two populations of house mice encountering different ecosystems and therefore are thought to have different demands of dietary response. Looking for adaptations in detoxification abilities I conducted a population based comparison of Cytochrome P450 (Cyp450) genes. These genes encode for detoxification enzymes and have already been shown to harbor an important source of adaptations to cope with xenobiotic compounds in different organisms. Clear indication for positive selection on three Cyp450 genes was found; both, selection on cis-regulatory elements was evident, as well as changes on protein level. Notably all three genes showed signs for selection in one of the investigated populations. Furthermore the affected genes are located on different chromosomes, supporting independent selective events within this gene family. This strongly indicates that the respective population evolved genetic responses to specific dietary compounds. However investigation of a priori chosen genes can only respond to previously held ideas, but cannot identify ‘unpredicted’ genes. Detection of previously unidentified or unsuspected genes that contribute to adaptation can be achieved by systematically screening the entire genome. Thereby whole chromosomes are scanned for ‘valleys’ of reduced heterozygosity (selective sweeps), a characteristic pattern caused by positive selection. In this case no a priori assumptions concerning the potential importance of genes or chromosomal regions are made before the scan is started; the genes are selected ‘bottom-up’. In the second part of the study, I present a genome screen for selection in different house mouse populations. Since the detection of polymorphic variants requires testing multiple individuals for several populations, a complete genome scan requires usage of a large number of markers. First a newly established method is described which facilitates high throughput analysis of microsatellites. A next generation sequencing based approach using the 454 technology was established which allows processing thousands of microsatellite loci simultaneously. I show that the obtained results are reliable and that the novel approach is a useful alternative to standard procedures. The above described sweep signatures are modified by several parameters such as the recombination rate and the selection coefficient. To reveal deeper insights into the basic parameters of positive selection and detection of chromosomal regions which might be target sites for selection, a genome screen was conducted including different populations of two house mouse subspecies (M. m. musculus and M. m. domesticus). I used the newly established method to investigate approximately 1,000 microsatellite markers on chromosome 19 in all populations. A detailed analysis of the candidate loci, identified by single comparisons, revealed results on the frequency of selective sweeps, and the putative origin of selected variants. Significant deviations of the sweep regions from the neutral state are statistically supported. Based on these results, I calculated that there was at least one positive selection event per 70 generations in each lineage. Furthermore, since only two sweeps indicate a broader sweep size than 80 kb, I conclude that positive selection is generally driven by alleles providing weak beneficial impact. Investigation of subspecies specific sweeps revealed shared signatures of selection between spatially and genetically distinct populations. This strongly indicates that beneficial mutations are potentially shared even among separated entities.