3dLME hangs when -mask option is included

AFNI Message Board
1

Hi AFNI experts,

I am running 3dLME on high resolution data that only covers a portion of the brain. I am using the -mask option to reduce the time and resources it will take, using this command:


3dLME -prefix Rest_LME -model "Cond*Side" \
-ranEff "~1" -SS_type 3 \
-jobs 6 \
-mask ./groupmask_all+tlrc \
-num_glt 6 \
-gltLabel 1 'Threat m Safe' \
-gltCode 1 'Cond : 1*T -1*S' \
-gltLabel 2 'LmR' \
-gltCode 2 'Side : 1*L -1*R' \
-gltLabel 3 'TmS in L' \
-gltCode 3 'Cond : 1*T -1*S Side : 1*L' \
-gltLabel 4 'TmS in R' \
-gltCode 4 'Cond : 1*T -1*S Side : 1*R' \
-gltLabel 5 'LmR in Safe' \
-gltCode 5 'Side : 1*L -1*R Cond : 1*S' \
-gltLabel 6 'LmR in Threat' \
-gltCode 6 'Side : 1*L -1*R Cond : 1*T' \
-dataTable \
Subj Cond Side IPIP_Extravers IPIP_Agreeable IPIP_Conscient IPIP_Intellect BIS_TOTAL BAS_RS BAS_Drive BAS_SS BAS_TOTAL EC_tot SC_tot PolAffil CESD_Total \
RSS_TOTAL Sens2Reward Sens2Punish SPIN_TOTAL BAI_TOTAL CSRI_TOTAL RSRI_TOTAL IUS_TOTAL PSWQ_TOTAL STAITrait_TOTAL_MRsess InputFile \
2202 T L 16 15 13 11 14 17 12 11 40 13 8 1 8 30 7 1 1 3 54 40 46 43 34 /bigraid-03/LS/Data/7T_BNST/afni/subs/2202/rest/results_restT_MNIb2/LBNST_Tzmap+tlrc \
2202 T R 16 15 13 11 14 17 12 11 40 13 8 1 8 30 7 1 1 3 54 40 46 43 34 /bigraid-03/LS/Data/7T_BNST/afni/subs/2202/rest/results_restT_MNIb2/RBNST_Tzmap+tlrc \
...

When I run this command, the output gets to:


Reading input files now... 
Reading input files: Done! 

If the program hangs here for more than, for example, half an hour, 
kill the process because the model specification or the GLT coding 
is likely inappropriate. 

[1] "Great, test run passed at voxel (75, 133, 112)!" 
[1] "Start to compute 225 slices along Z axis. You can monitor the progress" 
[1] "and estimate the total run time as shown below." 
[1] "10/06/16 11:06:36.369" Loading required package: snow Package snow loaded successfully!

And then does not progress, even after a few hours. However, if I comment out the -mask option, the program moves on after about 15 minutes and starts going through slices, with output like:


Z slice 1 done: 10/06/16 12:35:25.228
...

I’ve checked the mask that I am using - it is in the same space as my data, and is made up of only ones and zeros.

Anyone know what I could do to get the -mask option working?

Thanks!

2

That does seem weird. I can take a close look at the situation if you upload all the input files plus the 3dLME script.

3

I have uploaded a tar.gz with the script and data to:

https://panthers-my.sharepoint.com/personal/wsp_uwm_edu/_layouts/15/guestaccess.aspx?guestaccesstoken=qWzrNwE6tfyirJyk287TU4ygjGSFf0U9qBaV5keuAGg%3D&docid=172d1faadea9f4125a724b2c8a027ccbd&rev=1

The output from running the script with and without the mask option are also included in this folder (For the one without the mask option, I aborted once I saw that the slices were being processed, as it was only a test run).

Let me know if another format is better for you.

Thanks

Walker

4

For anyone who comes across this thread in the future, Gang Chen let me know that the script worked fine when he tried it, suggesting that the problem may have something to do with the computing cluster I am using.

Thanks again for looking into that!