[caret-users] Applying deformation map from Macaque atlas to Colin atlas

[David Van Essen]

Terry,

Several comments.

On Sep 17, 2008, at 5:53 PM, Terry wrote:

> Hi,
>
> I'm now trying to deform some borderproj files from the Macaque F99UA1
> atlas to the Colin atlas.  I first obtained a deform_map file by
> applying a spherical deformation, using a particular set of landmarks.
> Then, when I try to apply the deform_map file to the F99UA1 borderproj
> file, the process runs for a short while, and then I get the error
> message "Error
> deformed_deformed_to_Human.colin.L_Standard-Hum- 
> Monk_Macaque.F99UA1.L.SPHERE.STD.03-03-09.34132.coord:
> contains a different number of nodes".  I checked, and indeed this  
> file
> has 71785 nodes, rather than 34132.  This problem did not occur when I
> used the standard landmarks found in the Tutorial 3 directory, but it
> seems to me that if there was something wrong with the landmark files
> the deform_map file would not have been generated.  I'd appreciate any
> suggestions you may have.

1) Here's one possibility:

i) In the Spherical Surface Deformation: Deformation tab: you left the  
default Deform Individual to Atlas and Atlas to Individual
ii) you then tried to use the deformation_map file that maps Colin  
data to F99 rather than the separate deformation_map file (generally  
in a different directory) that maps F99 data to Colin.

See if you can track down the other deform_map and let me know if it  
now works.

2) I will be the first to admit that keeping track of these complex  
mappings from one surface to another is difficult, AND that we have  
not (yet) done an optimal job of setting up the file names so as to  
minimize confusion.  Improving this is on our to-do list.

3) Unless you have a compelling reason for using Colin as your human  
atlas, I would strongly encourage you to switch to our PALS-B12 atlas,  
which has numerous advantages.  Likewise, unless you are completely  
glued to the 34132-node mesh for the F99 macaque, I would strongly  
encourage you to switch to the 'standard-mesh' (73,730-node) version  
of F99.

The advantages of PALS-B12 and standard-mesh surfaces for monkey and  
human are numerous.  Many of them are articulated in the 'Caret - the  
Basics' tutorial (sept 06).  The tutorial document is in
http://sumsdb.wustl.edu/sums/archivelist.do?archive_id=6602379&archive_name=Caret_Tutorial_Oct06.pdf
and the dataset is in
http://sumsdb.wustl.edu/sums/directory.do?id=6585200&dir_name=CARET_TUTORIAL_SEPT-06

Part 4 describes monkey-human comparisons using standard-mesh F99 and  
PALS surfaces and includes an updated set of interspecies landmarks  
compared to what we previously used on Colin.

Available deformation map files allow your existing data on Colin and  
on native-mesh F99 to be transferred accurately to PALS-B12 and  
standard-mesh F99, respectively.

I hope this helps.

David

>
>
> Terry Sewards
> _______________________________________________
> caret-users mailing list
> caret-users at brainvis.wustl.edu
> http://brainvis.wustl.edu/mailman/listinfo/caret-users