-------- Original Message --------
Subject: Re: quantifying differences
Date: Tue, 24 Mar 2009 05:40:13 -0700 (PDT)
From: Guillaume COLOMBEAU <[email protected]>
To: [email protected]
References: <[email protected]>
Hi Anouchka,
i can't answer for IMP, but for Morphologika:
after procruste analysis, your data are geometrically split: size
informations are given by the centroid size (depends on calculation,
represents a mean of euclidean distances from each landmark to
centroid), and shape informations are given by the set of procruste
coordinates (resulting from translation, rotations AND scaling using
CS). In Morphologika, these last coordinates are used in PCA. The result
deals with shape, and you can eventually plot a principal component
versus CS graphic.
hope it helps
bye
guillaume
2009/3/24 morphmet <[email protected]
<mailto:[email protected]>>
-------- Original Message --------
Subject: quantifying differences
Date: Tue, 24 Mar 2009 03:55:57 -0700 (PDT)
From: JACQUIER Anouchka <[email protected]>
To: [email protected] <mailto:[email protected]>
<[email protected] <mailto:[email protected]>>
Hi everyone!
I'm still working on my 3D models of fishes. I have 68 homologous
landmarks on them and I have the correct format to enter my data in
IMP and Morphologika.
I'm trying now to compare my specimens. I wanted to show that two
species would be closer morphologically (eventhough separated
genetically = convergence case) than they are to their close
genetically related parents.
I wanted to compare their CS and Procrustes distance, but because
they are not the same size it gets difficult. I have centroid size
which are very different eventhough my specimens are really similar
morphologically.
Do you have any idea what kind of statistical test I could use and
which program is required. And do you know how to eliminate the
effect on size and only keep the effect of shape in my calculation,
which is really all that matters in my case.
I hope you will understand this rather complicated e-mail and hope
also that someone will be able to help me!
Thanks
Anouchka
Anouchka Jacquier
Master Student
Switzerland
e-mail: [email protected]
--
Replies will be sent to the list.
For more information visit http://www.morphometrics.org
--
Replies will be sent to the list.
For more information visit http://www.morphometrics.org
