[R-sig-phylo] problems with variance calculation in nuc.div

Emmanuel Paradis Emmanuel.Paradis at ird.fr
Thu Jan 19 04:31:19 CET 2012


Hi Paulina,

I don't have a copy of Nei (1987) with me, but it seems that the 
original formula is in Tajima (1983, Genetics, 105:437) which agrees 
with the one you attached with your message. I've corrected this in 
pegas: it'll be in the next release. Many thanks for reporting this.

Best,

Emmanuel

Paulina Ossa wrote on 18/01/2012 09:04:
> Hi,
> I ´ve been using the function nuc.div from the package pegas obtaining
> very high values for variance. I check my results with Arlequin, and I
> found serious differences between the variance results, the formulas
> used by both programs are different. unfortunately I dont have access to
> the original reference.
> below are both formulas, in red both terms in the equation that cause
> troubles:
>
> ARLEQUIN3.5:
>
>
>
> nuc.div
>
> var <- (n + 1) * ans/(3 * (n + 1) * dim(x)[2]) + 2 *
> (n^2 + n + 3) * ans/(9 * n * (n - 1))
>
> please, I need to know which is the right one, as I pointed above, i
> dont have access to Nei 1987 book.
>
> hoping you can help me,
> sincerely
>
> Paulina Ossa
> Phd Student
> PUC
> Santiago, Chile
>
>
>
> _______________________________________________
> R-sig-phylo mailing list
> R-sig-phylo at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-phylo

-- 
Emmanuel Paradis
IRD, Jakarta, Indonesia
http://ape.mpl.ird.fr/



More information about the R-sig-phylo mailing list