I have recieved personal communication from a kind group user explaining that fluctuating asymmetry is calculated from differences within individuals and so doesn't need phylogenetic correction.
So ignore my question! Thanks to those who viewed it anyway. On Wednesday, 1 April 2015 15:01:13 UTC+1, Alex Marshall wrote: > > Hello Morphomet group, > I am trying to conduct comparisons of the structure of fluctuating > asymmetry (FA) in squamate skulls in MorphoJ, both to FA structure in other > taxa and to symmetric shape covariance within the same taxa. I am using > matrix correlation with Mantel test for the comparisons. Normally, I > believe it is correct to use the Ind*Side covariance matrix produced > directly from Procrustes ANOVA for the matrix correlations. But I am also > trying to do this with independent contrasts, which is where confusion > starts. > > To create a phylogenetically-independent covariance matrix of FA I have > been mapping FA scores of shape data after Procrustes ANOVA onto a > phylogeny and then constructing a covariance matrix from the independent > contrasts. > > What confuses me is: > 1. What FA scores are and is it meaningful to construct covariance > matrices from them? > 2. Why Morphoj gives options to construct covariance matrices using > symmetric or asymmetric components from the independent contrasts of FA > scores? > (I have tried both and found they are different) > 3. Is the asymmetric component of shape (in MorphoJ) directional or > fluctuating asymmetry? In one paper (Fig. 6), > <http://www.jstor.org/stable/10.1086/650372> Klingenburg states it is the > latter, whereas I always assumed it is the former > > I hope this all makes some sense! > > Best, > > Alex Marshall > MSci Biology student > University College London > [email protected] > > -- MORPHMET may be accessed via its webpage at http://www.morphometrics.org To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].
