R help - Jan 2004 - Analyzing dendrograms

Ladies and Gentlemen,

As Johan Lindberg points out, the documentation for handling dendrograms
is sparse....Does anyone know who is responsible for or working on 
development of tree methods and objects? I've written a couple of scripts   
for my own use to translate between parenthetical (A(B(CD))) or binary  
A00 B10 C11 D11 tree formats and cluster objects in R, but as an 
inexperienced programmer have had difficulty integrating them with the 
existing framework--I'm trying to write some resampling routines and 
sensitivity tests for tree objects.

Perhaps this question should go to the r-devel list instead....

Yours,
Walton Green
> Date: Tue, 06 Jan 2004 12:49:59 +0100
> From: Johan Lindberg <johanl at kiev.biotech.kth.se>
> Subject: RE: [R] Analyzing dendograms??
> To: <r-help at stat.math.ethz.ch>
> Message-ID: <5.2.0.9.0.20040106122638.00badeb8 at
kiev.biotech.kth.se>
> Content-Type: text/plain; charset="iso-8859-1"; format=flowed
> ...dendograms/heatmaps/hclust in R in a proper way I ask myself, is 
> there a way of doing this in R? or is this a limitation of the 
> functions available today in the R programming language. I know there 
> are some functions like cutree etc but the documentation is really, 
> really sparse. Are there any tutorials out there of how to do these 
> things? or should one turn to alternative programs like MEV from TIGR?
> / J

>>>>> "Walton" == Walton A Green <walton.green at
yale.edu>
>>>>>     on Wed, 7 Jan 2004 10:41:03 -0500 (EST) writes:
    Walton> Ladies and Gentlemen,

    Walton> As Johan Lindberg points out, the documentation for
    Walton> handling dendrograms is sparse....Does anyone know
    Walton> who is responsible for or working on development of
    Walton> tree methods and objects? I've written a couple of
    Walton> scripts for my own use to translate between
    Walton> parenthetical (A(B(CD))) or binary A00 B10 C11 D11
    Walton> tree formats and cluster objects in R, but as an
    Walton> inexperienced programmer have had difficulty
    Walton> integrating them with the existing framework--I'm
    Walton> trying to write some resampling routines and
    Walton> sensitivity tests for tree objects.

Well, the R-core team is responsible for some...
at least, some of us (mainly Fritz Leisch, then me) have defined
and used the "dendrogram" class.

where  help(dendrogram) says quite clearly

  >> The code is still in testing stage and the API may change in the
  >> future.

Then, there's the much more general S4 classes in Bioconductor's 
"graph" package (which has a maintainer as every addon package...)

    Walton> Perhaps this question should go to the r-devel list
    Walton> instead....

yes.  There you could also elaborate a bit about what you
wanted, what you did, and where your difficulties were.

    Walton> Yours, Walton Green

      >> Date: Tue, 06 Jan 2004 12:49:59 +0100
      >> From: Johan Lindberg <johanl at kiev.biotech.kth.se>
      >> Subject: RE: [R] Analyzing dendograms??

      >> ...dendograms/heatmaps/hclust in R in a proper way I ask
      >> myself, is there a way of doing this in R? or is this a
      >> limitation of the functions available today in the R
      >> programming language. I know there are some functions
      >> like cutree etc but the documentation is really, really
      >> sparse. Are there any tutorials out there of how to do
      >> these things? or should one turn to alternative programs
      >> like MEV from TIGR?

      >> / J

Martin Maechler <maechler at stat.math.ethz.ch>
http://stat.ethz.ch/~maechler/
Seminar fuer Statistik, ETH-Zentrum  LEO C16	Leonhardstr. 27
ETH (Federal Inst. Technology)	8092 Zurich	SWITZERLAND
phone: x-41-1-632-3408		fax: ...-1228			<><