Table 4

Monte-Carlo simulations of the response to selection under the null hypothesis of absence of de novo mutations

nP

nH

estimates

othera

% simulations with linear response

Average responseb

P value


100

100

Late MBS

add

0.08

0.67 (0.43-0.99)

0.019*

100

60

Late MBS

add

0.12

0.46 (0.25-0.74)

0.320ns

100

60

Late MBS

dom

0.28

0.49 (0.23-0.76)

0.317ns

100

60

Early F252

add

0.24

-0.35 (-0.57 - -0.16)

0.068ns

100

20

Late MBS

add

0.33

0.22 (0.08-0.38)

0.046*

100

10

Late MBS

add

0.51

0.13 (-0.02-0.28)

0.006**

20

20

Late MBS

add

0.06

0.49 (0.27-0.78)

0.288ns

20

10

Late MBS

add

0.18

0.31 (0.13-0.55)

0.220ns

20

5

Late MBS

add

0.32

0.18 (0.04-0.36)

0.030*


a add = additivity within and between loci. dom = dominance of the most favourable allele. b 95% confidence interval is given between brackets.

Each simulation is defined by an initial number of polymorphic loci nP, an initial number of heterozygous loci nH and the experimental population from which the initial genetic variance was estimated. Corresponding values of initial heritabilities are given in (Table 3). % simulations with linear response is the fraction of the 500 runs for which the model that minimizes AICc in the segmented regression is the one with a breakpoint at G7 or after, indicating a linear response to selection. Average response is the average response to selection computed as in (3) for each run and averaged over the 500 runs. P value contains the percentage of simulations for which the average response to selection is lower than the one observed in the corresponding experimental population. When nH is too high, the simulated response to selection is always higher than the observed one. When nH is too low, the simulated response to selection is always lower than the observed one.

Durand et al. BMC Evolutionary Biology 2010 10:2   doi:10.1186/1471-2148-10-2

Open Data