[R-sig-eco] Interpreting large clusters

[Brian Knaus]

Hi Phil,

I'd start by asking if you can categorize your species in some sort of  
meaningful manner.  If so you can color code the labels of your  
dendrogram.  The package 'ape' includes 'tip.color=...' as a parameter  
in plot.phylo().  Create a vector of colors or numbers and use that to  
parameterize tip.color.

In regards to viewing the graphic I'd create a pdf.  This allows you  
to parameterize the paper size with width=7 and height=7 (units are  
inches).  You can parameterize this to height=14 or height=28 or other  
large values.  This may be printed over several pages or you can just  
scroll over it on your monitor.

Good luck!

-- 
Brian J. Knaus, Ph.D.
Department of Botany and Plant Pathology
Oregon State University
2082 Cordley Hall
Corvallis, OR 97331-2902
http://oregonstate.edu/~knausb


Quoting Phil Novack-Gottshall <pnovackg at westga.edu>:

> Hi all,
>
> I have a related question concerning cluster analysis of large data  
> sets.  In my case, the matrix is reasonably small for R to work  
> with, but I have so many species (~2000) that it is not possible to  
> read labels on the resulting dendrogram.  I imagine that using an  
> ordination is a preferable method in this case, but I was wondering  
> whether anyone had any recommendations for producing a very large,  
> but still readable dendrogram.  (I've tried increasing the window  
> size and shrinking cex.text, but this still isn't sufficient.)
>
> Cheers,
> Phil
>
>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>   Phil Novack-Gottshall                        pnovackg at westga.edu    
> Assistant Professor
>   Department of Geosciences
>   University of West Georgia
>   Carrollton, GA 30118-3100
>   Phone: 678-839-4061
>   Fax: 678-839-4071
>   http://www.westga.edu/~pnovackg
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> _______________________________________________
> R-sig-ecology mailing list
> R-sig-ecology at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
>