Show all posts by user
Dear AFNI users-
We are very pleased to announce that the new AFNI Message Board framework is up! Please join us at:
https://discuss.afni.nimh.nih.gov
Existing user accounts have been migrated, so returning users can login by requesting a password reset. New users can create accounts, as well, through a standard account creation process. Please note that these setup emails might initially go to spam folders (esp. for NIH users!), so please check those locations in the beginning.
The current Message Board discussion threads have been migrated to the new framework. The current Message Board will remain visible, but read-only, for a little while.
Sincerely,
AFNI HQ
History of AFNI updates
Hi Naveed,
That's strange... What do the following commands return (run them from the directory where you ran @SUMA_Make_Spec_FS):
afni -version
ls -l ./*
cheers,
z
by
ziad
-
AFNI Message Board
I am a little confused about the usefulness of the 6 parameters you mention.
3dcalc can be used to create a cubical ROI by making special use of the x, y, z variables or (i,j,k if you want to use voxel indices).
Here is how to create a box of dimensions: 100, 80, 60 mm and centered at (10, -3, -10)mm RAI . For a template grid, I used the TT_N27+tlrc volume in AFNI's bin directory:
3
by
ziad
-
AFNI Message Board
Hi PT,
No but those atlases are quite large and AFNI might be loading too many of them. There are ways to trim this by hiding some of the large atlases that you don't need from AFNI's sight (move them out of abin/ for now into some other directory). The better solution involves setting your preferences to load just a few of those atlases even if you have a ton available to AFNI. But
by
ziad
-
AFNI Message Board
Hi Nic,
The program estimates the delay between each voxel time series and the reference time series you used. The references you compare have slightly differing shapes but most importantly, have different time offset.
Since the program looks for positive delays, I suggest you do the following. Start from a square wave that starts a couple of TRs earlier than your actual stimulus timing squar
by
ziad
-
AFNI Message Board
Hi Naveed,
As Peter suggested, those files are not volume files and cannot be written directly into BRIK format. You can map cortical thickness values into the volume, but that would be a separate step with various options that you must consider.
If you have not done so already, you should run @SUMA_Make_Spec_FS -GIFTI -sid YOUR_SUBJECT
on the output of FreeSurfer. This will bring the thi
by
ziad
-
AFNI Message Board
Hi Anthony,
There were a couple of problems with the dataset. For starters, the orientation was off. In the axial view, the anterior direction was towards the bottom of the image viewer. Given that your dataset orientation was LPI, I changed it to RAI figuring it was unlikely that you have an orientation of LAI which would be a left handed coordinate system but I can't tell for sure. In a
by
ziad
-
AFNI Message Board
Hi Anthony,
Upload away and send me an email when you do so. I'll take a look.
cheers,
z
by
ziad
-
AFNI Message Board
Hi Asali,
Hard to tell without the dataset in hand. Does the mprage have a lot of neck coverage? Anything different in the appearance of this dataset compared to what 3dSkullStrip has worked well on in the past?
cheers,
z
by
ziad
-
AFNI Message Board
Hi Vinai,
There are a few options. See how THD_dataset_mismatch() is used in 3dinfo.c . That should clue you in on a variety of mismatching flavors, and there are many!
cheers,
z
by
ziad
-
AFNI Message Board
tsliu Wrote:
-------------------------------------------------------
> Hi Ziad,
> Thanks for your reply. I didn't know all this is
> possible, that's great news. I've seen the
> suma.pdf document. I thought it was lecture
> slides, but I guess it also serves the purpose of
> a manual? If so we shall study it carefully.
It is meant for the bootcamp, but
by
ziad
-
AFNI Message Board
Hi Cecil,
We don't have such a capability for plugout_drive at this moment though there has been a few requests to that effect. For the moment, your only option is to run 3dmerge (-1clust* options) on the dataset with the clustering parameters of your choice and then open the resultant dataset with AFNI for image making.
cheers,
Ziad
by
ziad
-
AFNI Message Board
Hello Kai,
3dQwarp is not for surface creation, it performs non linear registration to bring a pair of volumes into alignment.
As to afni_restproc.py, it is not a script that we maintain, contact its author Rayus Kuplicki (rkuplicki@laureateinstitute.org) if you don't hear from him on this message board soon.
cheers,
Ziad
by
ziad
-
AFNI Message Board
Hello Taosheng,
Sorry for the delay, but it is Gang Chen's fault! You can certainly do all of 1. and all of 2.
The manual/documentation is mostly in suma.pdf . The outline goes something like this:
1- Create the surfaces, with FreeSurfer say.
2- Run AFNI's: @SUMA_Make_Spec_FS -sid Festus -GIFTI
That would import FreeSurfer's surfaces, parcellations (volume and surface
by
ziad
-
AFNI Message Board
Hi Josh,
If your AFNI is current, then the problem is on our side. There was a problem with the build on linux machines that I did not notice until now. This will be fixed by the morning, but in the meanwhile, if you used @update.afni.binaries -d to update your code, you will find a copy of your previous 3dSkullStrip under abin/auto_backup (with abin/ being the directory where your AFNI binaries
by
ziad
-
AFNI Message Board
Hi Julia,
We have been using the matlab function RetroTS.m to create regressors of no interest from physiological recordings. The program @afni_proc.py can use RetroTS's output as part of the resting state pre-processing pipeline. If you don't have matlab, see if you can use the stand-alone version per the instructions here.
cheers,
Ziad
by
ziad
-
AFNI Message Board
Hello folks,
I think that is because it is voxelizing the surface, and that takes a while for high res data. I will look into turning this off unless the users ask for it. Just to be sure though, it is running forever or just quite slowly? If it is going on forever could you send me a sample command so that I can reproduce this? If I also need data I can let you know later.
cheers,
z
by
ziad
-
AFNI Message Board
Hi Isaac,
Hard to tell without experimenting what would do best. Usually the options I try first would be -shrink_fac, -init_radius. You can also try to uniformize the volume with 3dUnifize before running 3dSkullStrip.
Lastly, consider running the program in -visual mode to get a feel for what is going on.
cheers,
ziad
by
ziad
-
AFNI Message Board
Hi Peter,
That should be good, but you should also be able to read the .gii file directly as a dataset into SUMA too, much like you did with the .1D file.
cheers,
Ziad
by
ziad
-
AFNI Message Board
Hi Nick,
I'd hate to make 3dAllineate attempt alignments that are potentially quite far off, especially when the transform to MNI space should be available somewhere in that pipeline. If you just can't get your hands on it, you can go with the parameters you set if they work well with the data you have. I would expect you might have some problems every now and then and the process as
by
ziad
-
AFNI Message Board
Hi Peter,
I second Nick's suggestions. To go between AFNI/SUMA and FreeSurfer your best best is to go via NIFTI for volumes and GIFTI for surfaces and surface-based data.
cheers,
z
by
ziad
-
AFNI Message Board
We have had weird rendering problems reported before. I suggest you update your graphics card driver first and preferably use the card vendor's version. Even when SUMA displays OK locally, it may not display well when run remotely. In most cases a driver (and Mesa/OpenGL library) update usually takes care of the problem.
Note that SUMA will be painfully slow when run remotely, a serious buz
by
ziad
-
AFNI Message Board
Hi Emily,
> I was wondering about a few things. I know that
> you can adjust the seed radius. Is it possible to
> visualize the seed on the afni gui, so I can see
> where the seed radius extends too and do a visual
> inspection?
Not interactively, but you can easily do so with the program 3dUndump with something like:
3dUndump -master ~/abin/MNI_caez_N27+tlrc -xyz -ori
by
ziad
-
AFNI Message Board
Hi,
I am a little confused about what you're trying to do. If you're using FreeSurfer parcellations, the ROIs you extract will be in subject space though not necessarily in alignment with your data.
You also mentioned MNI 152 ROIs. If you want those resampled to original space then you can do so after you estimate the transform of each subject's data to MNI_152 space. You can
by
ziad
-
AFNI Message Board
Hi Isaac,
The bug is now fixed. I hope to rebuild tonight, but if you want the fix before then, you can add this one line:
DSET_mallocize(dset); DSET_load(dset);
after the line with
DSET_quiet_overwrite(dset);
in SUMA_SegFunc.c, line 661 (line number could be slightly different)
I never ran into this before because I do not explicitly set the byte order. Why is it that you do so
by
ziad
-
AFNI Message Board
Hi Isaac,
That's some serious debugging effort! I'll take a look at this tomorrow. I have not had time to look into this at all yet.
Thanks for the tenacious effort, hopefully I'll manage to reproduce the problem here otherwise I'll have to ask you to test.
cheers,
z
by
ziad
-
AFNI Message Board
Hi Mingbo,
-nocor is not discontinued, and -use_mgz does not result in noticeable differences compared to the default.
I recommend you use -NIFTI instead. I don't know what else to add in terms of clarification beyond what is in the -help output for these options. Is there anything in particular that needs clarification?
cheers,
Ziad
by
ziad
-
AFNI Message Board
Don't bother with the patch, try grabbing afni_src.tgz again, it should have the fix in it now.
cheers,
z
by
ziad
-
AFNI Message Board
Hi Alex,
Sorry for that, I don't know how I failed to notice this!
The function call in SUMA_path_optimize.c line 433 should be:
SDO = SUMA_CreateSegmentDO( N_n, oriented, NodeBased, Stipple,
Label, idcode_str, Parent_idcode_str,
type_not_set, NULL, LineWidth, LineCol,
NodeId, NULL, n0,
by
ziad
-
AFNI Message Board
Greetings,
In SUMA versions postdating this message GIFTI surfaces will be assumed to be in LPI, rather than RAI, in keeping with the GIFTI standard.
This change means that GIFTI surfaces created by older versions of SUMA, will be flipped along the R--L and A--P axes when loaded with new versions of SUMA. As a result, mixing old SUMA GIFTI surfaces with new SUMA binaries or vice versa, wil
by
ziad
-
AFNI Message Board