Blog

Saludos from the 17IHIW

by Monterey Bay, California.

One thing that usually surprises Europeans is how much bigotry and creationism are spread in this paradise of science and technique that the USA are supposed to be.

Looks like astrology or lovetalk but in an opposite mode, that is, we don’t really read and use horoscopes, but we talk as if the subject deserved some credit. I start to believe that it is a general feature of this country’s culture. Here I’ve seen a lot of talk about population genetics but then no use of it.

If a sample does not conform with Hardy-Weinberg expectations then the sample frequencies are at most sample frequencies and certainly not *population* frequencies. If a sample is in Hardy-Weinberg equilibrium then it makes no sense to say that the population is composed of two or three different sub-populations, as, if it was the case, the Wahlund effect would ensure deviation from Hardy-Weinberg expectations: you can’t have simultaneously two exclusive situations.


2 comments on “Saludos from the 17IHIW

    • Kudos for the question.
      The answer is no, but there are two different kinds of no.
      On one side HWE is a robust condition from a statistical point of view, which means that samples not fulfilling the HWE conditions don’t show departure from HWE expectations. On the other side there is what is called pseudo-random mating populations where HWE expectations are met yet we can show clear violations of the HWE assumptions.
      In practice, if HWE is rejected, clearly there is a problem, if it doesn’t then assume there are no (strong) problems. Of course, HWE applies to each single locus, so some consideration for multiple testing must be used. It is mostly in this last point that there is disagreement between practitioners.

Join the discussion, leave a comment