[Bioperl-l] questions about the bioperl module Bio::PopGen::Statistics

[Warren W. Kretzschmar]

Hi Jun,
First, in case you haven't seen the howto page, here it is:
<http://www.bioperl.org/wiki/HOWTO:PopGen>

Now to address your question.  Unfortunately I can only give you a
short, inadequate answer: pi is the average number of pairwise
differences between the sequences of n individuals and is also an
estimator of the scaled mutation rate (theta).  Pi assumes a
wright-fisher model, which is not directly based on the concept of
DNA.  Theta is not the per base-pair mutation rate.  Instead theta = 2
* M * mu, where M is the effective population size and mu is the per
base-pair mutation rate.  If you want to learn more, I suggest reading
the lecture notes that can be found at
<http://www.stats.ox.ac.uk/~myers/mathgenteaching.html>, especially
the notes for lectures 5-7.

Cheers,
Warren

On Wed, Jun 1, 2011 at 1:30 PM, lvu.jun <lvu.jun at gmail.com> wrote:
> Hi, there,
> I am trying to calculate the population genetics parameters such as pi using the bioperl module Bio::PopGen::Statistics. But I found that the method only requires the input of the marker genotype of every individuals for the population. I don't know why the module does not take the DNA sequence length into consideration when calculating the pi value. According to the definition of the pi value, besides the polymorphic sites, we also need the monomorphic sites that should be incorporated in the denominator when doing the calculation. Is it right? therefore I'm confused about the module, who can tell me why it can correctly calculate the pi value only with the marker(polymorphic) genotype?
> Another question, if I want to calculate the pi value using the sliding window along the genome, how can I do this using the Bio::PopGen::Statistics module?
> Thanks for your help!
> Yours sincerely,
> Jun
>
> Chinese Academy of Sciences
>
> 2011-06-01
>
>
>
> lvu.jun
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/bioperl-l
>