Dear AFNI experts,
We have done block design and retinotopy (Only 2 wedges) experiments in one scanning session. I have done GLM analysis on block design data in afni, and now I’d like to run retinotopy analysis on the wedge scans and project the phase map on the surface files to draw V1 boundaries. The anatomical image (obtained in the same session) is segmented with MIPAV. I use a script which generates SUMA surface files based on MIPAV output but not the -spec files. So, I can’t run @RetinoProc. I have tried using 3dRetinoPhase, but it crashes with the following error:
** ERROR: Bad rpud->Fresp !
Fatal Signal 11 (SIGSEGV) received
3dRetinoPhase main
Bottom of Debug Stack
** Command line was:
3dRetinoPhase -clw Wedge1_r.nii.gz Wedge2_r.nii.gz -prefix Ret -Tstim 36 -nwedges 1 -pre_stim 12 -phase_estimate FFT
Given that I don’t have any eccentriciy data (rings) and spec files, I’d like to know whether it is possible to do retinotopy analysis with afni.
Any help is greatly appreciated.
Kind regards,
Khazar Ahmadi
You can create spec files with quickspec. See the help for that program for some examples.
Thanks a lot for your prompt reply.
I could create spec files using quickspec but when I run @Retinoproc, it doesn’t generate any .niml files due to missing SurfVol. Is there a way other than @SUMA_Make_Spec to create a surface volume?
In addition, during the alignment step when align_epi_anat.py is running, skull is removed from the anatomy however the T1 image is already skull stripped. How can I change the script in a way that it skips 3dskullstrip.
Many thanks in advance for your help.
Best regards,
Khazar Ahmadi
The SurfVol usually needed can be any dataset that has been transformed from the volumetric space of your other datasets to the space of the surfaces. One advantage of that is to create surfaces once and apply data to those surfaces from other days. If you are doing everything in only one native space, just use any existing volumetric dataset. The surface programs, including suma, only look for affine transformations in the header and apply those to move surfaces to volumes.
Many thanks for your reply.
Still, the script doesn’t generate any .niml files due to missing r/l h.smoothwm.asc files. Is there a way to generate such a file?
Any help is greatly appreciated.
Kind regards,
Khazar Ahmadi
Looking at this, it seems the script is using the .asc format. I’m guessing you are using the .gii (GIFTI) format surfaces instead. I’ve modified the script to check for the file type. I haven’t checked it with real data, so if you would like to try it out, it is here:
https://afni.nimh.nih.gov/pub/sscc/staff/glend/@RetinoProc
Please let me know how it works for you.