-------- Original Message -------- Subject: Re: About relabeling reflection Date: Sat, 28 Oct 2006 06:18:53 -0700 (PDT) From: [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] To: [email protected]
Dear Pablo, Chris Klingenberg is the best person I know to ask this question. It seems to me that in his 2002 paper, he provides guidelines on how to do if you're analysing symmetric structures but you're not interested in asymmetry. To my knowledge, to have asymmetries sometimes is not great for the GPA and visualization. However, you might have noticed that most geometric morphetric analyses of symmetric structures not focussing on asymmetry do not even mention this issue. Most of the time it does not probably make any (appreciable) difference in results of your tests. Possibly, you can do a simple test: remove the asymmetric component (very easy to do following Chris' protocol), compute a matrix of Procrustes shape distances on the symmetric dataset, do the same with the original dataset (including asymmetries), compute a matrix correlation. If you get a very high correlation (0.99 something), that means that positions of your specimens in the shape space are about the same (or differences are proportional) regardless of asymmetries. It may be tricky if you have a strong directional asymmetry. In that case, the matrix correlation will be high (I guess), but the position of the specimens in the shape space may differ (being more or less proportional). Whether this, and other potential issues with asymmetries, matter to you, it will depend on your questions. Check my suggestions with people who have experience on this kind of analyses. I have very little experience and might be wrong. Another good paper I know on analysis of aymmetric structures has been published this year or last year by people in Vienna and other colleagues. I do not have the reference nearby but you can ask Philipp Mitteroecker, one of the authors. Good luck Andrea >Dear morphometers, > >After reading a series of papers dealing with the issue of bilateral symmetry >and the treatment of asymmetric variation, my conclusion is >that the relabeling reflection, suggested by works such as those of >Mardia et al (2000), Bookstein and Mardia (2003) and Klingenberg et al. >(2002), is the method most convenient and parsimonious with the theory >of shape variation. > >However, I would like to turn to you for advise regarding the convenience of >applying such a method when analyzing large series of >bilaterally symmetric structures for the study of geographic variability >and evolution. My interest is not directly on the effects of asymmetry, >but overall variability in the size and shape of the skull. So my >question is how important is to eliminate the effects of asymmetry from >the shapes of interest? Is this treatment for data something that you >will suggest as critical? Or is it a step I could perhaps obviate, >assuming that the degree of asymmetry is small enough, perhaps after a >pilot test on asymmetry variance? > >I will be grateful with any advice you could provide me. I am a biologist who >loves morphometrics, working amidst sequencers and DNA >laboratories, so you could imagine my longing for discussing >morphometric issues with more experienced colleagues on the subject. > >Thanks > >Pablo > >Pablo Jarrin Grad. Student >Dept. of Biology >Boston University > >-- >Replies will be sent to the list. >For more information visit http://www.morphometrics.org > > > Dr. Andrea Cardini Hull York Medical School The University of York, Heslington, York YO10 5DD, UK PLEASE, GET IN TOUCH WITH ME BY EMAIL BEFORE USING THE ADDRESS ABOVE. E-mail: [EMAIL PROTECTED], [EMAIL PROTECTED] http://www.york.ac.uk/res/fme/people/andrea.htm ------------------------------------------------------------ Sei stanco di girare a vuoto? Con il nuovo motore di ricerca Interfree trovi di tutto. Vieni a trovarci: http://search.interfree.it/ Lo Staff di Interfree ------------------------------------------------------------ -- Replies will be sent to the list. For more information visit http://www.morphometrics.org
