[Camino-users] problems using conmat

[Longchuan Li]

Hi, Camino users
I have encountered an issue using camino command 'conmat' and would appreciate your help on it:
I generated a brain atlas covering cortical and subcortical regions with values ranging from 33 to 122. The intensity values in the atlas are not continuos. When I tried to get the connectivity matrix, I got the follow error:
my command: cat BedpostxTracksV2i5ct60ci4.Bfloat | conmat -targetfile brain_atlas.nii.gz -outputroot ./conmat_

Error:
Aug 19, 2015 12:12:48 PM tractography.TractSource <init>INFO: reading data from standard inputException in thread "main" java.lang.IllegalArgumentException: Illegal index -1 is < 0 or exceeds index of last point        at tractography.Tract.pathLengthFromSeed(Tract.java:277)        at tractography.StreamlineConnectivityGraph.processTract(StreamlineConnectivityGraph.java:281)        at tractography.StreamlineConnectivityGraph.processTracts(StreamlineConnectivityGraph.java:201)        at apps.ConnectivityMatrix.writeConnectivityGraph(ConnectivityMatrix.java:367)        at apps.ConnectivityMatrix.execute(ConnectivityMatrix.java:138)        at apps.EntryPoint.main(EntryPoint.java:287)
However, when I manually drew a few cortical ROIs in a brain mask and used 'conmat', I got sensible results. So, the problem is definitely somewhere in the brain atlas I used. So far, I have tried the following:
(1) I loaded the brain atlas to matlab and saved it to nifti again using "save_avw", I got the program run, but all values are zero. The main difference in the matalab-saved brain atlas is that matlab "save_avw" will not save coordinate information in the brain atlas (will give a eye mat). So I assume the issue is related to the coordinates. Because there is no coordinate information in matlab-written brain atlas, there might be a mismatch in space between tracts and mask and therefore generated all zero results. 
(2): discontinued intensity values in the brain atlas is not an issue.  
Many thanks!Longchuan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.nitrc.org/pipermail/camino-users/attachments/20150819/28c0560b/attachment.html>