Phenotypic evolution from genetic polymorphisms in a radial network architecture

Arnaud Le Rouzic¹^,2 , Paul B Siegel³ and Örjan Carlborg¹^,4

¹Linnaeus Centre for Bioinformatics, Uppsala University, Box 598, SE-75124 Uppsala, Sweden

²Centre for Ecological and Evolutionary Synthesis, Department of Biology, PO 1066 Blindern, 0316 Oslo, Norway

³Virginia Polytechnic Institute and State University, Department of Animal and Poultry Sciences, Blacksburg, VA 24061-0306, USA

⁴Department of Animal Breeding and Genetics, Swedish University of Agricultural Sciences, Box 7023, SE-750 07 Uppsala, Sweden

author email corresponding author email

BMC Biology 2007, 5:50doi:10.1186/1741-7007-5-50


Published:	14 November 2007

Additional files

Additional file 1:

Effect of the presence of extra low-effect QTLs on the expected dynamics. The average selection response of the network itself is compared with that of the network plus six small-effect independent loci. Increasing the number of small additive QTLs clearly leads to a more linear response that fits better with the experimental data.

Format: PDF Size: 165KB Download file

This file can be viewed with: Adobe Acrobat Reader

Additional file 2:

Impact of the sampling effect. The genotype-phenotype map was evaluated from a finite F₂population and the estimates of genotypic values are necessarily subject to sampling effects. The most sensitive genotypes are likely to be fully homozygous, because they are less frequent in the F₂.

Format: PDF Size: 254KB Download file

This file can be viewed with: Adobe Acrobat Reader

Additional file 3:

Effect of the missing genotypes estimation on the overall dynamics. The table summaries the four missing genotypes and their estimates following three different methods: taking the average of the population (624 g), estimating from the mean of the neighbor genotypes and estimating from the regressions of Figure 2. The figure presents the resulting dynamics, with the same parameters as Figure 3B. The differences are of the same order of magnitude as the stochastic differences between the repetitions, and the method to estimate the missing genotypes appears to have almost no influence on the results.

Format: PDF Size: 242KB Download file

This file can be viewed with: Adobe Acrobat Reader

Additional file 4:

Impact of the simplification of the model. Simulations were simplified compared with the actual selection experiment. One of these simplifications was to consider an even and constant number of males and females selected each generation; the other was to consider that the population size was constant in both lines. Simulations show that the results remain consistent when these hypotheses are relaxed.

Format: PDF Size: 183KB Download file

This file can be viewed with: Adobe Acrobat Reader