The last comment is
important. It is not sufficient to simply apply any clustering method and just
look at the resulting dendrogram. One needs to use some method to determine how
well the dendrogram actually fits the data. One cannot tell be just looking at
the dendrogram. For example, UPGMA is not expected to work well on data unless
there actually are clusters in the data (there even has to be nested clusters of
clusters) and thus it is not always an appropriate method.
---------------------- F. James Rohlf,
John S. Toll Professor, Stony Brook University The
much revised 4th editions of Biometry and Statistical Tables are now
available: http://www.whfreeman.com/Catalog/product/biometry-fourthedition-sokal http://www.whfreeman.com/Catalog/product/statisticaltables-fourthedition-rohlf Please
consider the environment before printing this email From:
morphmet_modera...@morphometrics.org
[mailto:morphmet_modera...@morphometrics.org] Dear Rodrigo, On Mon, Nov 19, 2012 at
6:41 PM, <morphmet_modera...@morphometrics.org>
wrote: Date: Mon, 19 Nov 2012 10:46:37 -0500 Subject: RE: UPGMA: Euclidean and Procrustes
distances Dear Andrea, Oyvind, and
Philipp, From: morphmet_modera...@morphometrics.org
[morphmet_modera...@morphometrics.org] Euclidean distances are equal to Procrustes distances in
tangent space, so they are close enough to the original Procrustes
distances. Best, Philipp Am
15.11.2012 um 20:32 schrieb morphmet_modera...@morphometrics.org: Dear
morphometricians, I would greatly appreciate your
input on this matter: The context: I am studying 9
populations of a small rodent species and I wanted to represent morphological
relationships between populations using a UPGMA cluster
analysis. The problem: I used the
mean population shapes (Procrustes coordinates) to do this analysis in PAST
using Euclidean distances and got a nice tree with bootstrap values, but there
is no option for using Procrustes distances in this package. I get the same
topology using Procrustes distances with R, but I can’t get the bootstrap
values on the nodes of the tree. The
questions: 1 - Are UPGMA analyses using
Euclidean distances and Procrustes distances
equivalent? 2 - Is there any package in which I
could simply enter the Procrustes distances and get a UPGMA tree with bootstrap
values? Thank you in
advance, Rodrigo ___________________________________ Dr. Philipp
Mitteroecker carlosce2...@gmail.com Departamento de Botânica, Ecologia e Zoologia -
DBEZ
----- Forwarded message from "F.
James Rohlf"
Date: Tue, 27 Nov
2012 11:55:21 -0500
From: "F. James Rohlf"
Reply-To: ro...@life.bio.sunysb.edu
Subject: RE: UPGMA: Euclidean and Procrustes distances
To:
morphmet@morphometrics.org
Sent: Tuesday, November
27, 2012 2:17 AM
To: morphmet@morphometrics.org
Subject: Re:
UPGMA: Euclidean and Procrustes distances
----- Forwarded message from morphmet_modera...@morphometrics.org
-----
Date: Sun, 25 Nov 2012 16:54:27 -0800
From: morphmet_modera...@morphometrics.org
Reply-To:
morphmet_modera...@morphometrics.org
Subject:
Re: UPGMA: Euclidean and Procrustes distances
To: morphmet@morphometrics.org
----- Forwarded message from Carlos Eduardo R D Alencar
-----
Date: Sat, 24 Nov 2012 11:36:40 -0500
From: Carlos Eduardo R D
Alencar
Reply-To: Carlos Eduardo R D Alencar
Subject: Re: UPGMA:
Euclidean and Procrustes distances
To: morphmet@morphometrics.org
Good
question. That was one of the doubts that I had in mind and were well clarified.
My use was to check similarities between groups of crabs in different species
and locations.
Well, in a second time you might want to check what the best
method of biding grouping (clustering). I would like to leave a tip,
according to analyzes already performed by me (in several aspects)
morphological, ecological (communities) and etc. ... UPGMA method was not very
well. My assessment is based on two coefficients (cophenetic correlation
coefficient and aglomeration coefficient). Maybe you want to test them before
using the UPGMA.
Cheers
Carlos
-----
Forwarded message from Rodrigo Lima -----
From: Rodrigo Lima
Reply-To: Rodrigo Lima
To: "morphmet@morphometrics.org"
Thank you very much for your answers, they totally clarified my
doubts.
Best wishes,
Rodrigo
Sent:
November 18, 2012 1:40 AM
To: morphmet@morphometrics.org
Subject:
Re: UPGMA: Euclidean and Procrustes
distances
-----
Forwarded message from Philipp Mitteröcker -----
Date: Fri, 16 Nov 2012
11:18:49 -0500
From: Philipp Mitteröcker
Reply-To: Philipp Mitteröcker
Subject: Re: UPGMA: Euclidean and Procrustes distances
To: morphmet@morphometrics.org
----- Forwarded message from
Rodrigo Lima -----
Date: Thu, 15 Nov 2012 12:21:28 -0500
From:
Rodrigo Lima
Reply-To: Rodrigo Lima
Subject: UPGMA: Euclidean
and Procrustes distances
To: "morphmet@morphometrics.org"
----- End forwarded
message -----
Department of Theoretical Biology
University of
Vienna
Althanstrasse 14
A-1090 Vienna, Austria
Tel: +43 1 4277 56705
Fax: +43 1 4277 9544
email: philipp.mitteroec...@univie.ac.at
homepage:
http://theoretical.univie.ac.at/people/mitteroecker
-----
End forwarded message -----
----- End forwarded message
-----
--
Carlos Eduardo R. D.
Alencar
Biólogo, Msc.
Bolsista
CAPES - Programa de Pós-graduação em Ecologia - PPGE
Grupo de Estudos em
Ecologia e Fisiologia de Animais Aquáticos - GEEFAA [www.geefaa.com]
Universidade Federal do Rio Grande do Norte -
UFRN
-----
End forwarded message -----
----- End forwarded message
-----
-----
End forwarded message
-----
- UPGMA: Euclidean and Procrustes distances morphmet_moderator
- Re: UPGMA: Euclidean and Procrustes distances morphmet_moderator
- Re: UPGMA: Euclidean and Procrustes distances morphmet_moderator
- Re: UPGMA: Euclidean and Procrustes distances morphmet_moderator
- RE: UPGMA: Euclidean and Procrustes distances morphmet_moderator
- Re: UPGMA: Euclidean and Procrustes distances morphmet_moderator
- RE: UPGMA: Euclidean and Procrustes distances morphmet_moderator
