[adegenet-forum] what does 'd = ' mean in pca plots

Jason Paul Joines jason at joines.org
Wed Dec 6 15:49:41 CET 2017


         Interesting.  Then it seems that 'd = a' only describes 
properties of the plot and says nothing about the analysis.

         Thanks for digging that out of the code.


Jason
===========



---------------------------------Original-Message---------------------------------
> Date: Wed, 6 Dec 2017 09:10:00
> From: Roman Luštrik <roman.lustrik at biolitika.si>
> To: Jason Paul Joines <jason at joines.org>
> Cc: adegenet-forum at lists.r-forge.r-project.org
> Subject: Re: [adegenet-forum] what does 'd = ' mean in pca plots
> 
> Following the code, I can trace it to ade4::scatter.grid (https://github.com/cran/ade4/blob/d571403a7ed08b2649c44901014aeb9e480ac1b0/R/scatterutil.R#L506).
> `xaxp` is a vector of coordinates and number of intervals between tick marks (see ?par). The code takes second and first coordinate and divides it by the number of intervals. And `a`, which is `d = a`, appears to be the minimum of this for x and y. Sorry for not being less technical.
>
> HTH,
> Roman
>
>
> ----
> In god we trust, all others bring data.
>
> ----- Original Message -----
> From: "Jason Paul Joines" <jason at joines.org>
> To: adegenet-forum at lists.r-forge.r-project.org
> Sent: Wednesday, December 6, 2017 5:00:51 AM
> Subject: [adegenet-forum] what does 'd = ' mean in pca plots
>
> adegenet Users,
>
>         I used the scatter function to plot the glPca of a genlight object
> as shown in tutorial-genomics.pdf.  The plot includes a small inset
> barplot of Eigenvalues in the bottom left corner.  The top right corner
> has the text "d = n" where n is a number that differs from one plot to the
> next.  What does "d = n" represent?
>
>
> Thanks,
>
> Jason
> ===========
> _______________________________________________
> adegenet-forum mailing list
> adegenet-forum at lists.r-forge.r-project.org
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/adegenet-forum
>
>


More information about the adegenet-forum mailing list