The question how much non-neutral evolution is involved in shaping a genome has been a constant theme since Kimura (1983) proposed "The neutral theory of evolution". In the meantime however several studies found evidence for non-neutral directional selection working on the base-composition of certain genes.
So far no study has been done trying to detect traces of directional selection in a genome wide survey.
Taking Drosophila melanogaster as a model organism and using microsatellites as neutral markers we are now carrying out such a survey. The strategy we use is to analyze microsatellite variability over the whole genome. Microsatellite variability is expected to be reduced if adjacent to a site that has been fixed in a population. To be fixed means that only the one allele that confers an advantagous mutation reaches a frequency of one whereas all other alleles are lost. This process is called a selective sweep. Together with the advantagous allele neutral sites in its flanking region are getting fixed as well during the selective sweep. They are said to "hitchhike" with the positive mutation. The size of the hitchhiked region depends on the effective population size, the recombination frequency and the selection coefficient of the positive mutation. For this genome screen we use natural populations from Europe and Africa.
Major questions we adress with this experimental
A certain pattern, first found in D.melanogasterbut subsequently confirmed in different species as well, is a correlation of nucleotide diversity and recombination rate.
Two completly different models of selection are invoked to account for this pattern: selective sweeps (positive directional selection - see above) and background selection (negative selection). The model of background selection describes the recurrent removal of deleterious mutations from the genome. Similarly, as for the hitchhiking model, linked neutral loci are removed from the genome together with the negative mutations resulting in reduced neutral polymorphism. Both models predict the largest reduction of variability in regions of reduced crossing over and therefore both could account for the pattern described above.
One way, however, to descriminate between the two models of selection is to compare levels of genetic variability on the X chromosome and on autosomes. X chromosomes are expected to have greater variability than autosomes under the background selection model and vice versa.
Comparing X chromosomes and autosomes from D.melanogasterwe recently found that levels of variability are quite different in african (ancestral) compared to European populations (colonized about 10.000 years ago) revealing a pattern consistent with background selection in Africa and indicating selective sweeps in European populations. Consistent with that we recently found two loci where microsatellite varibility is significantly reduced only in European populations. However further investigations concerning the relative contributions of hitchhiking and background selection in African populations seem more appropriate.
Kauer, M., Dieringer, D. & Schlötterer, C. (2002). Natural selection against immigrant genotypes suggest incipient speciation between African and non-African D. melanogaster. submitted
Kauer, M., Dieringer,D. & Schlötterer, C. (2002). A microsatellite variability screen for colonization associated positive selection in the genome of D. melanogaster. in prep.
Harr, B., Kauer, M. & Schlötterer, C. (2002). Hitchhiking mapping: a population-based fine-mapping strategy for adaptive mutations. P.N.A.S., in press.
C., Kauer, M. & Dieringer, D. (2002). Allele excess at neutrally evolving
microsatellites and the implication for tests of neutrality. submitted.
Kauer, M. & Schlötterer, C.: "Evidence for positive selection in non-African populations of D. melanogaster", Conference of the European Society for Evolution (ESEB 2001, Aarhus/Denmark). (poster).
M & Schlötterer, C.: "Putatively selected regions in the genome
of Drosophila melanogaster identified in a screen of microsatellite
variability", Conference of the Society for Molecular Biology and Evolution
(SMBE 2002, Sorrento/Italy). (poster).