;
so we expect, in this case, that the frequency of the A allele at
equilibrium will be 0.76/(0.08+0.76)=0.90, and the sickle cell gene should
have a frequency,
at equilibrium, of 0.10.
We can check our answer by working backwards,
and using selection coefficients to calculate equilibrium
frequency: if we were given the above selection coefficients, the
equilibrium gene frequency should be;
so we expect, in this case, that the frequency of the A allele at
equilibrium will be 0.76/(0.08+0.76)=0.90, and the sickle cell gene should
have a frequency,
at equilibrium, of 0.10.
If we did all the above calculations while minimizing rounding errors, we would actually get p*=0.0908, the frequency we initially estimated above under "testing whether the data conforms to Hardy Weinberg."
Calculator note: once again, don’t round down any number if you can store it in memory at higher precision. By rounding to two decimal places, we confused ourselves into thinking that we got a different answer for p* when in fact it should have been the same!(It’s a little silly to estimate the selection coefficients from the genotype frequencies and then go on to estimate what the equilibrium should be using those same selection coefficients, as we have done here. It would be an example of circular thinking. We had to make the assumption of equilibrium in order to estimate the selection coefficients in the first place. However, it was worth doing the exercise to see that you can calculate the results in either direction).