curious asymmetry in mutation frequencies

[Rupert Mazzucco]

I am observing a curious asymmetry in simulated mutation frequencies with respect to selection coefficients in a symmetric 1-locus model. The attached plot shows median frequency differences from the starting frequency of 50% over 30 replicates at the symmetrically placed target position after 100 generations as a function of the absolute value of the selection coefficent. Why do negative selection coefficients cause a larger frequency change than the corresponding positive coefficients, is this supposed to be so?

Still on SLiM 4.0.1 (the copr repo seems to be defunct?)

Many thanks

Rupert

[Philipp Messer]

Hi Rupert,

In the Wright-Fisher model with selection, the expected absolute change in allele frequency shouldn't actually be exactly the same for positive and negative s. Even in the haploid model, we have E[Delta_x] = sx(1-x)/(1-sx). Thus, for a biallelic locus with alleles A and B, when A has selection coefficient s over allele B, this is not exactly equivalent to B having selection coefficient -s over A. I know this might feel a bit unsatisfying, but it's what it is in the WF model, and I think it explains your observation unless I misunderstood something. 

Best,

Philipp 

--
SLiM forward genetic simulation: http://messerlab.org/slim/
---
You received this message because you are subscribed to the Google Groups "slim-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to slim-discuss...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/slim-discuss/84525168-e741-4e59-a414-a2fa228fe9fdn%40googlegroups.com.

[Rupert Mazzucco]

No, that is probably it. Many thanks!

Rupert