Missing aroma outputs

Hello everybody,

I´m new to fmriprep and hope someone can advice me here. I used fmri prep on the same data twice with two different modifications - one time using the freesurfer and sync and the other time not. In both cases I added --use-aroma. However only with freesurfer enabled aroma outputs were generated. Is aroma only working with freesurfer? I couldn’d find any related post, so I guess its rather my code being wrong…

Thanks a lot for helping!!
Anna

The code with freesurfer was:

nohup docker run -i --rm \

-v /usr/local/freesurfer/license.txt:/opt/freesurfer/license.txt:ro \

-v /data/anna/memory/fmriprep/data:/data:ro \

-v /data/anna/memory/fmriprep/output/out:/out \

-v /data/anna/memory/fmriprep/output/work:/scratch \

poldracklab/fmriprep:1.2.1 \

/data /out/out \

participant \

–output-space T1w template fsnative fsaverage5 \

–participant_label 8 9 \

–use-aroma \

–use-syn-sdc \

–nthreads 4 \

–write-graph \

-w /scratch


The code without freesurfer was:

nohup docker run -i --rm \

-v /usr/local/freesurfer/license.txt:/opt/freesurfer/license.txt:ro \

-v /data/anna/memory/fmriprep/data:/data:ro \

-v /data/anna/memory/fmriprep/output/out:/out \

-v /data/anna/memory/fmriprep/output/work:/scratch \

poldracklab/fmriprep:1.2.1 \

/data /out/out \

participant \

–output-space T1w template \

–participant_label 8 9 \

–use-aroma \

–nthreads 6 \

–fs-no-reconall \

–omp-nthreads 4 \

–write-graph &

Hi @anna,

I don’t see any reason that you shouldn’t be getting AROMA outputs without FreeSurfer enabled. Were there any errors at the bottom of your reports when you ran without FreeSurfer?

I do see that you don’t seem to be using -w /scratch in your --fs-no-reconall run, which will result in your working directory falling entirely within the Docker container, which might run up against disk space issues. (If not, then it’s no big deal; you just can’t reuse the working directory when re-running.)

Chris

Hi Chris,

thanks a lot for the quick reply! I will add the -w /scratch in the --fs-no-reconall command just in case I just realized I did not use it when posting the code here.

I didn’t get any errors in the html report nor in the command window. In the command window there were many warnings though. I hope it´s ok to ignore those… But looking at the command window output again I guess I realized my misstake. I accidently put not space after the number 9 in the participant _label. Thats why I only got this one participant processed. But I did no think that this affected aroma too. However it did: Some participants were not found: 9–use-aroma … I’m sorry for bothering you only because I did not pay enough attention!

Thanks a lot!!
Anna

Hi Chris,

One quick question now that I know what happened: I do not have the aroma data now for a few patients. Can I simply rerun it using the aroma flag correctly with leaving all the data in the directories? And will this save runtime compared to removing the data from the output directories prior to running it with the aroma flag?

Thanks a lot!
Anna

Preserving or removing data from the output directories won’t have any significant effect on run time, but preserving the working directory will.