sliding semilandmarks and save Procrusted coordinates

[Pablo Fisichella]

Dear All

I have a couple of basic questions. I am conducting an GM analysis using 3D landmarks and semilandamrks. I define  semilandmarks using the function define.sliders() and I used the fuction proc <- gpagen() in order to call the Procrustes adjusted coordinates and  save them to perform analyses in another sotfware. However, when I compared the consensus configuration without slinding or slinding using Procrustes distance and bending energy they appear quite similar (except when I used the Bending energy the consensus configuration looks different) and when I tried to save the proc$coords the same values appear for each specimen independently if I did not used the sliding or use the Procrustes distance or bending energy. So, I am confused if really I am sliding semilandmarks and how I proceed to saving the Procrustes coordinates (with the sliding semilandmarks). Here I attach some screenshots to show my example. In each figure appear the GPA in the R console and the Procrustes coordinates of the first individual (AL2001aUM2). For instance in Fig 1 (without sliding) and 3 (sliding using using Procrustes distance) the consensus configuration is similar and the Procrustes coordinates are the same. In Fig 2 (sliding using the Bending energy) the cosensus configuration is relatively different regarding Figs 1 and 3 but the Procrustes coordinates of the first specimen (AL2001aUM2)) are the same as in Fig 1 and Fig 2.

thanks so much for any suggestion and please excuse my long e-mail 

workflow 1 (not slinding)

1. x<-read.morphologika(file.choose())#call data

2. gpagen(x$coords)# here the sliding was not performed

3.  proc <- gpagen(x$coords)#extracting the Procrustes coordinates

4. a<-(proc$coords)# assign to the object a the Procrustes coordinates

5. a# calling the Procrustes coordinates

workflow 2 (slinding using Procrustes distance and bending energy)

1. x<-read.morphologika(file.choose())#call data

2. a<-read.csv(file.choose())# loading the curveslide.csv file

3. gpagen(x$coords,a)# Using Procrustes Distance for sliding or gpagen(x$coords,a, ProcD=FALSE)#Using bending energy for sliding

4.  proc <- gpagen(x$coords)#extracting the Procrustes coordinates

5. a<-(proc$coords)# assign to the object a the Procrustes coordinates

6. a# calling the Procrustes coordinates

[Mike Collyer]

Pablo,

In steps 2 and 3 of workflow two you vary the sliding method but do not assign the results as an object. In step 4 you assign the results but do not use a sliding method.  This is why, I believe, you are getting results that appear to be the same. 

Mike

Sent from my iPhone

--
You received this message because you are subscribed to the Google Groups "geomorph R package" group.
To unsubscribe from this group and stop receiving emails from it, send an email to geomorph-r-pack...@googlegroups.com.
To post to this group, send email to geomorph-...@googlegroups.com.
Visit this group at https://groups.google.com/group/geomorph-r-package.
To view this discussion on the web, visit https://groups.google.com/d/msgid/geomorph-r-package/692b9322-0a7e-4133-a900-656309a0410d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

<Fig 1..jpg>

<Fig 2.jpg>

<Fig 3.jpg>