open-discussion > compiling code for shapeAnalysisMANCOVA
Showing 1-25 of 26 posts
Display:
Results per page:

1   2   Next >
Jul 22, 2015  10:07 PM | Lara Foland-Ross - Stanford University
compiling code for shapeAnalysisMANCOVA
Hello Spharm-pdm experts, 

With many thanks to you, I finally am at a stage where I can begin testing for group differences in hippocampal shape (yay!). As I've run everything via command line thus far, I'd like to continue doing with shapeAnalysisMANCOVA which I understand I need to install.

I am a novice when it comes to compiling. Are there instructions available on how to download, compile and install shapeAnalysisMANCOVA?

Secondly, can you please tell me what other programs must be installed for shapeAnalysisMANCOVA - is it just ITK (ver4.7) and VTK (ver5.1)?

Many thanks,
Lara
Jul 23, 2015  10:07 AM | D J
RE: compiling code for shapeAnalysisMANCOVA
Hi Lara,

Under linux I use the pre-compiled version in the 1.12 release. No need to compile it yourself.
I think you can even copy just the shapeanalysisMANCOVA executable into your existing spharm folder from the later release and it still works.

Regards,
David.
Jul 23, 2015  01:07 PM | Martin Styner
RE: compiling code for shapeAnalysisMANCOVA
Agree with David. shapeanalysisMANCOVA has not changed in a while, so it should be the same in most recent releases.
Martin
Jul 23, 2015  04:07 PM | Lara Foland-Ross - Stanford University
RE: compiling code for shapeAnalysisMANCOVA
It doesn't appear that shapeAnalysisMANCOVA is included in the 1.13 and 1.5 releases for mac OS X. 

Do you suggest that I download another package version? I am running everything on Mac OSX 10.8.5.

Thanks,
Lara
Jul 24, 2015  01:07 PM | Martin Styner
RE: compiling code for shapeAnalysisMANCOVA
Hi Lara

Indeed we need to do a compilation for MacOS X. We don't have a 10.8 machine in the lab though, so I am not sure that is easily done (we run 10.10 in the lab). I think there is an option to compile things on the Mac such that they are backward compatible (I think...)

Do you have access to a linux box? This may solve this step very quickly.

Martin
Jul 24, 2015  01:07 PM | Beatriz Paniagua
RE: compiling code for shapeAnalysisMANCOVA
Hi Lara, 



That is correct, shapeAnalysisMANCOVA (SAM) is not at the moment available for MacOS. Sometime after the 1.12 version we decided to split apart SAM and SPHARM-PDM because SAM was no longer supported. Also, sometime after that we started offering MacOS compilations of SPHARM-PDM (when SAM was no longer a component of it).

I suggest using a Linux box, as Martin mentioned. We are working on developing a new statistical framework for SPHARM, but we are not quite there yet. SAM will have to be the statistical framework for SPHARM a little bit longer.


Sorry for the trouble. Please, let us know if you need help using a linux box.
Thank you,


Beatriz
Jul 24, 2015  03:07 PM | Lara Foland-Ross - Stanford University
RE: compiling code for shapeAnalysisMANCOVA
Hi Beatriz and Martin, 

I do not have access to a linux box unfortunately. Does this mean that mac users are not able to run ShapeAnalysisMANCOVA?

I'd be very happy to try and compile the source code myself, if it is available.

Thanks,
Lara
Jul 24, 2015  03:07 PM | Beatriz Paniagua
RE: compiling code for shapeAnalysisMANCOVA
Hi Lara,


Correct, mac users are not able to run ShapeAnalysisMANCOVA. Have you tried to use https://www.virtualbox.org/?
It is a linux box system developed by Oracle that can be used by MacOS users. You could install a fresh scientific linux OS in there (this is what we use at the NIRAL).

The shapeAnalysisMANCOVA code was ported to Github (https://github.com/NIRALUser/shapeAnalys...)  with the rest of the NIRAL tools, but I would like to highlight it is currently unsupported since we are trying to devise new shape statistic methods. On the other hand, we will be happy to provide feedback and help if you try to build the tool it, and you are welcome to try to debug and change the actual code.


Beatriz
Jul 28, 2015  09:07 PM | Lara Foland-Ross - Stanford University
RE: compiling code for shapeAnalysisMANCOVA
Hi Beatriz, Martin and David, 

Many thanks for your continued help and responses to my numerous questions.

I have successfully installed and am currently running Ubuntu on virtualbox. I have organized my data and am running the pre-compiled version of shapeAnalysisMANCOVA found in the v1.2 linux build that David recommended. I'm very excited to get this far but now have the following error message which I'm not able to trouble shoot on my own. Advice on how to proceed would be appreciated:

shapeAnalysisMANCOVA inputlist.csv --coulmnGroupTypes 0 --infileColumn 2

input file specified: inputlist.csv
filename: inputlist.csv numIndependent: 0
Num Subjects: 20
data in group_type 0 has been relabeled: 0 --> group A = -1 ; 1 --> group B = 1
#(A)= 10; #(B)= 10
terminate called after throwing an instance of 'std::out_of_range'
what(): basic_string::compare
Aborted (core dumped)

my inputlist.csv file is attached. I have double checked that the paths are correct and that the files are there. Any clues as to what I'm missing?

Of note, I 

Many thanks in advance,
Lara
Attachment: inputlist.csv
Jul 28, 2015  09:07 PM | Lara Foland-Ross - Stanford University
RE: compiling code for shapeAnalysisMANCOVA
Sorry - premature email sending. I meant to say that it may be worth noting that I created my vtk files on my mac, copied them over to the virtual machine, and then ran everything on the virtual machine. Not sure if this matters?

Many thanks again,
Lara
Jul 29, 2015  02:07 PM | Martin Styner
RE: compiling code for shapeAnalysisMANCOVA
vtk files are cross-platform, thus this should not matter
Martin
Jul 29, 2015  02:07 PM | Martin Styner
RE: compiling code for shapeAnalysisMANCOVA
2 things:
a) small typo in columnGroupTypes (though that does not make a differences as "0" is the default)
b) the csv is not a csv but a tsv (i.e it should be comma separated and not tab separated. 

Can you adjust your csv file and try again?

Martin
Jul 29, 2015  03:07 PM | Lara Foland-Ross - Stanford University
RE: compiling code for shapeAnalysisMANCOVA
Thanks for catching those! It works now! 

Thanks again,
Lara
Jul 29, 2015  04:07 PM | Lara Foland-Ross - Stanford University
RE: compiling code for shapeAnalysisMANCOVA
Hi again,

While no errors were printed from running the following command:

shapeAnalysisMANCOVA ../abnormalhead_procalign_inputlist.csv --columnGroupTypes 0 --infileColumn 2 -n 1000 --testColumn 0

The only output is a meta file for each participant of the second group (prefix to meta file is identical to prefix of vtk file for each subject in the second group).

My input list file is again attached. The exact text returned from the command is copy and pasted below.

Many thanks in advance for the continued help,
Lara

kristen@kristen-VirtualBox:~/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS$ sudo shapeAnalysisMANCOVA ../abnormalhead_procalign_inputlist.csv --columnGroupTypes 0 --infileColumn 2 -n 1000 --testColumn 0
 input file specified: ../abnormalhead_procalign_inputlist.csv
filename: ../abnormalhead_procalign_inputlist.csv numIndependent: 0
Num Subjects: 20


data in group_type 0 has been relabeled: 0 --> group A = -1 ; 1 --> group B = 1
#(A)= 10; #(B)= 10
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17011.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk -1 6.92081e-310
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17012.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk -1 6.92081e-310
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17013.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk -1 2.37152e-322
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17014.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk -1 2.37152e-322
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17015.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk -1 1.43808e-316
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17016.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk -1 4.94066e-324
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17018.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk -1 2.122e-314
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17019.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk -1 9.21755e+227
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17020.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk -1 4.74303e-322
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17010.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk 1 3.16202e-322
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17011.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk 1 1.43808e-316
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17012.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk 1 1.13635e-322
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17013.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk 1 2.122e-314
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17014.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk 1 2.1612e+185
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17015.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk 1 9.20546e+179
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17016.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk 1 3.66388e-307
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17018.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk 1 1.72632e-306
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17019.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk 1 4.79244e-322
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17020.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk 1 1.43805e-316
 -1 1.43809e-316
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17011.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17012.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17013.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17014.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk - .vtk

reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17015.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17016.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17018.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17019.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17020.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17010.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17011.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17012.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17013.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17014.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17015.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17016.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17018.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17019.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17020.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk - .vtk
reading Mesh  -   - kristen@kristen-VirtualBox:~/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS$
Jul 29, 2015  08:07 PM | Martin Styner
RE: compiling code for shapeAnalysisMANCOVA
The issue seems that it does not recognize all lines/surfaces, as the first mesh is not listed/read. If I remember correctly, the first line in the csv needs to be a header line:

#group, don't know, mesh
0,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17010.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk
0,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17011.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk
0,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17012.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk
0,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17013.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk
0,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17014.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk
0,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17015.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk
0,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17016.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk
0,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17018.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk
0,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17019.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk
0,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17020.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_procalign.vtk
1,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17010.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk
1,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17011.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk
1,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17012.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk
1,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17013.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk
1,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17014.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk
1,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17015.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk
1,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17016.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk
1,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17018.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk
1,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17019.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk
1,1,/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17020.1_Lhipp_surf_reorient_original_round2SPHARM_procalign.vtk
Jul 29, 2015  09:07 PM | Lara Foland-Ross - Stanford University
RE: compiling code for shapeAnalysisMANCOVA
Thank you Martin! That did the trick!

I've worked through a couple more issues but am now stumped with the following error message. Can you please advise?

No statistic type selected, defaulting to Hotelling.
Now running nonparametric tests:
Performing MANCOVA permutation testing.
Using 1000 permutations.
Completed MANCOVA permutation testing.
fdr thresh (MANCOVA) is = 0.002
Runtime was 0 minute(s) and 57 second(s).
terminate called after throwing an instance of 'std::out_of_range'
what():  basic_string::erase

Again, my input was
sudo shapeAnalysisMANCOVA abnormalhead_procalign_inputlist.csv --columnGroupTypes 0 --infileColumn 2 -n 1000 --testColumn 0

Thanks in advance,
Lara
Aug 3, 2015  01:08 PM | Martin Styner
RE: compiling code for shapeAnalysisMANCOVA
Hi Lara, 
mm, not sure, is there anything written out, surfaces generated? Runtime of about a minute seems awfully short, but possible if there aren't too many surfaces (as in your case).
Best regards
Martin
Aug 3, 2015  08:08 PM | Lara Foland-Ross - Stanford University
RE: compiling code for shapeAnalysisMANCOVA
Hi Martin,

A .meta file was produced for each subject, as well as the following files:

customLUT_DiffMagnitude.txt
customLUT_FDRP.txt
customLUT_RawP.txt
distances_scaled.txt
listfile_DiffMagnitude.txt
listfile_diffMesh.txt
listfile_mancovaBonferroniP.txt
listfile_mancovaFDRP.txt
listfile_mancovaRawP.txt
listfile_normProjections.txt
listfile_GLM_correctedMean.meta
listfile_meanAll_uncorrected.meta
listfile_meanA.meta
listfile_meanB.meta

I am again pasting in the exact command and output below, as well as attaching my input csv file. Any help is much appreciated!

Thanks,
Lara

kristen@kristen-VirtualBox:~/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS$ sudo shapeAnalysisMANCOVA listfile.csv --columnGroupTypes 0 --infileColumn 2 -g 1 -n 100 --hotelling
 input file specified: listfile.csv
filename: listfile.csv numIndependent: 0
Num Subjects: 20


data in group_type 0 has been relabeled: 0 --> group A = -1 ; 1 --> group B = 1
#(A)= 10; #(B)= 10
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17010.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk -1 6.90091e-310
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17011.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk -1 6.90091e-310
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17012.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk -1 2.122e-314
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17013.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk -1 2.42092e-322
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17014.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk -1 0
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17015.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk -1 1.1907e-321
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17016.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk -1 6.90091e-310
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17018.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk -1 6.90091e-310
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17019.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk -1 2.122e-314
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17020.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk -1 1.39351e-319
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17010.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk 1 0
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17011.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk 1 9.53547e-322
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17012.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk 1 6.90091e-310
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17013.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk 1 6.90091e-310
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17014.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk 1 2.122e-314
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17015.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk 1 1.56029e-317
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17016.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk 1 0
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17018.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk 1 7.16395e-322
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17019.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk 1 1.32612e-316
/home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17020.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk 1 6.90091e-310
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17010.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17011.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17012.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17013.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17014.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17015.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17016.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17018.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17019.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17020.1_Lhipp_surf_reorient_original_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17010.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17011.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17012.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17013.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17014.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17015.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17016.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17018.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17019.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk - .vtk
reading Mesh /home/kristen/Desktop/QA_TESTING/ABNORMAL_HEAD/STATS/17020.1_Lhipp_surf_reorient_abnormalhead_round2SPHARM_ellalign.vtk - .vtk
Using Hotelling statistics.
Now running nonparametric tests:
Performing MANCOVA permutation testing.
Using 100 permutations.
Completed MANCOVA permutation testing.
fdr thresh (MANCOVA) is = 0
Runtime was 0 minute(s) and 6 second(s).
terminate called after throwing an instance of 'std::out_of_range'
  what():  basic_string::erase
Attachment: listfile.csv
Aug 4, 2015  09:08 AM | D J
RE: compiling code for shapeAnalysisMANCOVA
Hi Lara,

Never come across that error yet from memory.

That said, I've found SPHARM to be somewhat tempermental. Couple of trouble shooting steps with the CSV file:
1. Avoid blank spaces (e.g. give your columns dummy headings)
2. Try making your inputfile paths in the first column, and putting all the data afterwards (and if first column in CSV is inputs and second column is group labels, then since SPHARM starts counting from zero change the command, so --infileColumn 0 --numGroupTypes 1 --columnGroupTypes 1 --testColumn 1 ). My own analysis have failed if the input files are in another column, although it throws a different error to yours. Can still try.

Hope this helps,
David.
Aug 5, 2015  05:08 PM | Lara Foland-Ross - Stanford University
RE: compiling code for shapeAnalysisMANCOVA
Hi David, 

Thanks for your suggestions. Unfortunately playing with the setup of the .csv file did not help. 

If anyone has other suggestions, I'm all ears! Otherwise, will look for other programs that can be used in place of shapeAnalysisMANCOVA.

Thanks again,
Lara
Aug 5, 2015  05:08 PM | Beatriz Paniagua
RE: compiling code for shapeAnalysisMANCOVA
Hi Lara,



Have you checked if you have p-values and the average meshes make sense?
You should have something like: [input_csvname]_uncorrected_All.meta/vtk can you look at those in Slicer?

Thanks,
Beatriz
Aug 5, 2015  09:08 PM | D J
RE: compiling code for shapeAnalysisMANCOVA
The other idea is to ensure the input meshes are all valid. You could try opening all of them in Slicer to ensure there's been no errors along the way. In particular, I believe failed spherical parameterisation can leave you with blank meshes (but still the VTK files are present) if you use the command line.
Aug 6, 2015  02:08 PM | Martin Styner
RE: compiling code for shapeAnalysisMANCOVA
Did you actually check the output whether the data that has been written out makes sense? it could be that it crashes very late and that you have all the data that you need.

Can you run it with 1000-10000 permutations and check out the p-value files and the mean meshes that are being written out?

Martin
Aug 6, 2015  10:08 PM | Lara Foland-Ross - Stanford University
RE: compiling code for shapeAnalysisMANCOVA
Hi Martin, 

Apologies for the delay in my reply. I took some time to reinstall virtualbox/Scientific Linux to see if that was an issue, but the problem persists.

I ran the command using 10,000 permutations and see one vtk with the following attributes: DiffMagnitude, DiffMagnitude_original, FDRP, FDRP_original, Normals, RawP, RawP_original. I also have four .meta files: listfile_GLM_correctedMean.meta, listfile_meanAll_uncorrected.meta, listfile_meanA.meta and listfile_meanB.meta. I can load the vtk file in ShapePopulationViewer.

Are there any inputs that I may be missing?

Thanks,
Lara
Aug 6, 2015  11:08 PM | Lara Foland-Ross - Stanford University
RE: compiling code for shapeAnalysisMANCOVA
Hi again, 

Quick update - I found that altering the command line to have "./listfile.csv" instead of "listfile.csv" helped move things along - with a new error message this time. Here's the files that resulted:

customLUT_DiffMagnitude.txt
customLUT_FDRP.txt
customLUT_RawP.txt
distances_scaled.txt
listfile.csv
listfile.csv_commandline.txt
listfile.csv_DiffMagnitude.txt
listfile.csv_diffMesh.txt
listfile.csv_GLM_correctedMean.meta
listfile.csv_mancovaBonferroniP.txt
listfile.csv_mancovaFDRP.txt
listfile.csv_mancovaRawP.txt
listfile.csv_meanAll_uncorrected.meta
listfile.csv_meanAll_uncorrected.vtk
listfile.csv_meanA.meta
listfile.csv_meanB.meta
listfile.csv_normProjections.txt
listfile_DiffMagnitude.txt
listfile_diffMesh.txt
listfile_GLM_correctedMean.meta
listfile_mancovaBonferroniP.txt
listfile_mancovaFDRP.txt
listfile_mancovaRawP.txt
listfile_meanAll_uncorrected.meta
listfile_meanAll_uncorrected.vtk
listfile_meanA.meta
listfile_meanB.meta
listfile_normProjections.txt
transformFiles

...and here's the error (see the very end):

No statistic type selected, defaulting to Hotelling.
Now running nonparametric tests:
Performing MANCOVA permutation testing.
Using 10000 permutations.
Completed MANCOVA permutation testing.
fdr thresh (MANCOVA) is = 0.0027
Runtime was 4 minute(s) and 22 second(s).
Mean_Difference_Min:-1.85594_Max:4.97653
Mean_Difference_Min:-1.85594_Max:4.97653
CreateMRML ./listfile.csv_GroupTest_MRMLscene.mrml -t -f ./transformFiles/TransRawP.tfm -n transRawP -l 1,0,0,0,1,0,0,0,1,10,0,-60 -m -f listfile.csv_meanAll_uncorrected.vtk -n RawP -p transRawP -as RawP -cc customLUT_RawP.txt -q -id RawP -lbl RawP -pos 21.1383,-12.6241,59.2907 -t -f ./transformFiles/TransFDRP.tfm -n transFDRP -l 1,0,0,0,1,0,0,0,1,10,0,-42.9618 -m -f listfile.csv_meanAll_uncorrected.vtk -n FDRP -p transFDRP -as FDRP -cc customLUT_FDRP.txt -q -id FDRP -lbl FDRP -pos 21.1383,-12.6241,42.2524 -t -f ./transformFiles/TransMeanOverlayRight.tfm -n transMeanOverlayRight -l 1,0,0,0,1,0,0,0,1,107.544,0,-25.9236 -m -f listfile.csv_meanA.meta -n MeanAOverlayRight -p transMeanOverlayRight -as MeanAOverlayRight -dc 1,0,0 -op 0.6 -m -f listfile.csv_meanB.meta -n MeanBOverlayRight -p transMeanOverlayRight -as MeanBOverlayRight -dc 0,0,1 -op 0.4 -t -f ./transformFiles/TransMeanOverlayLeft.tfm -n transMeanOverlayLeft -l 1,0,0,0,1,0,0,0,1,-87.5444,0,-25.9236 -m -f listfile.csv_meanA.meta -n MeanAOverlayLeft -p transMeanOverlayLeft -as MeanAOverlayLeft -dc 1,0,0 -op 0.4 -m -f listfile.csv_meanB.meta -n MeanBOverlayLeft -p transMeanOverlayLeft -as MeanBOverlayLeft -dc 0,0,1 -op 0.6 -q -id Overlays -lbl Overlays -pos 31.7074,-12.6241,19.1951 -t -f ./transformFiles/TransMeanRight.tfm -n transMeanRight -l 1,0,0,0,1,0,0,0,1,107.544,0,-8.88534 -m -f listfile.csv_meanA.meta -n MeanRight -p transMeanRight -as MeanRight -dc 1,0,0 -q -id Means -lbl Means -pos 31.7074,-12.6241,2.15689 -t -f ./transformFiles/TransMeanLeft.tfm -n transMeanLeft -l 1,0,0,0,1,0,0,0,1,-87.5444,0,-8.88534 -m -f listfile.csv_meanB.meta -n MeanLeft -p transMeanLeft -as MeanLeft -dc 0,0,1 -t -f ./transformFiles/TransDiffMagnitude.tfm -n transDiffMagnitude -l 1,0,0,0,1,0,0,0,1,10,0,8.15288 -m -f listfile.csv_meanAll_uncorrected.vtk -n DiffMagnitude -p transDiffMagnitude -as DiffMagnitude -cc customLUT_DiffMagnitude.txt -q -id Mean_Difference_Min:-1.85594_Max:4.97653 -lbl Mean_Difference_Min:-1.85594_Max:4.97653 -pos 42.2766,-12.6241,-20.9004 Error: Could not run CreateMRML:
No such file or directory
Error: Could not run CreateMRML:
No such file or directory

...as usual, advise on how best to proceed/troubleshoot would be much appreciated.

Thanks,
Lara

1   2   Next >