----- Forwarded message from Carmelo Fruciano -----
Date: Thu, 25 Oct 2012 10:07:57 -0400
From: Carmelo Fruciano
Reply-To: Carmelo Fruciano
Subject: Re: Ontogentic shape comparison
To: morphmet@morphometrics.org
morphmet_modera...@morphometrics.org ha scritto:
>
> ----- Forwarded message from "SANTOS SANTOS, JAVIER" -----
>
> Date: Wed, 24 Oct 2012 12:43:23 -0400
> From: "SANTOS SANTOS, JAVIER"
> Reply-To: "SANTOS SANTOS, JAVIER"
> Subject: Ontogentic shape comparison
> To: morphmet_modera...@morphometrics.org
>
> Hello morphometricians,
>
> I am comparing ontogenetic shapes in two species of Lake Victoria
> cichlids in MorphoJ (2D). I want to compare shape differences at each
> ontogenetic stage. My question lies in what would be the better option
> to compare the shapes visually (so for a illustration), and I would
> like to know which of these two options you think is best to do so:
> (1) Do a PCA in MorphoJ for each species at a given ontogenetic stage.
> Then compare the consensus configuration for each. Or (2) Do the same
> PCA and then compare both shapes at a determined scale factor (i.e.
> 0.1). Or (3) Do a combined PCA and then compare the shapes obtained
> with a negative scale factor (where one species is located) and with a
> positive scale factor (where the other species is located).
> Thanks in advance for your advice.
>
Dear Javier,
I'm not sure I understand. Do you want to plot at each stage the
difference between the two species as a shape change?
You could do a Generalized Procustes Analysis of everything together,
then compute mean Procustes residuals for each stage of each species
and finally produce deformation grids between pairs of interest (for
example in tpsSplin, if I remember).
Another way to do it could be, after the GPA, just use the
discriminant function routine in the program to get the graph for the
difference in mean shape between the relevant pairs.
If you, instead, use the PCA module in the program you will get, as
far as I know, only shape changes associated with a certain PCA while,
as far as I can tell, you wanna plot shape changes between species
spanning all the dimensions (not only the first PC).
I hope this helps,
Carmelo
--
Carmelo Fruciano
Post-doc - University of Konstanz - Konstanz, Germany
Honorary Fellow - University of Catania - Catania, Italy
e-mail c.fruci...@unict.it
http://www.fruciano.it/research/
----- End forwarded message -----
