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
It's been a while, but like Isaac I also used the 10.8_gcc Makefile with some modifications (pointing the dependencies to things installed by Homebrew instead of fink or to fit my own convoluted build schemes). As for dependencies...
Xm.h is provided by either openmotif or lesstif. I think I managed to get it to work with Homebrew.
brew install lesstif
But you may also be abl
by
Peter Molfese
-
AFNI Message Board
Hi Jared-
Even though there is no VBM pipeline, you can still use AFNI for the group analyses after you compute the necessary files using: Freesurfer, FSL-VBM, FSL_ANAT or SPM's VBM8
-Peter
by
Peter Molfese
-
AFNI Message Board
There isn't a way to do more than one overlay at a time. However, you can use 3dcalc or InstaCalc to do conjunction style maps. For instance you could take mapA and mapB and see the overlap or non-overlap using something like:
3dcalc -a mapA+tlrc -b mapB+tlrc -expr 'step(a-2.4)+2*step(b-2.4)' -prefix mapAB
where the value of 2.4 is some threshold of your t-statistic.
by
Peter Molfese
-
AFNI Message Board
Yes, you can do surface smoothing using the AFNI/SUMA tools. You'll want to look at SurfSmooth.
Note that typically you want to blur or smooth before running statistics, which often means projecting your data to the surface earlier in data processing. You might want to look at Example 8 of the afni_proc.py documentation.
by
Peter Molfese
-
AFNI Message Board
It's hard to tell what your afni_proc.py script contained. But assuming it's one of the canned examples with an align and tlrc block: The anatomical that the EPI is co-registered with is the anat_final.0050772+tlrc.HEAD. It will likely be at a different resolution than your EPI file (depending on your afni_proc.py command). You can verify this with a command like:
3dinfo -prefix
by
Peter Molfese
-
AFNI Message Board
Hi Richard,
What does:
ls -al /usr/local/Cellar/gcc/5.2.0/lib/gcc/5/libgomp.1.dylib
return?
The reason for the question is that you can actually link (ln) to a nonexistent file:
ln -s /some/random/path/to/no/file ~/Desktop
So it's possible that your dylib is in a different location.
by
Peter Molfese
-
AFNI Message Board
If you're using the "install GCC" option, you still need to link libgomp to the directory that 3dAllineate wants it to be in. Something like this should work:
ln -s /usr/local/Cellar/gcc/5.2.0/lib/gcc/5/libgomp.1.dylib /usr/local/lib/libgomp.1.dylib
Though make sure that the file really exists. If you don't see it, verify that you installed GCC with the "--with-a
by
Peter Molfese
-
AFNI Message Board
My current workaround is now posted to the original thread:
by
Peter Molfese
-
AFNI Message Board
One work around for getting AFNI to work without disabling SIP:
0. If you upgraded from Mac OS X 10.10 (or earlier), reinstall XQuartz.
1. Install XCode via Mac App Store:
2. Install Homebrew: Link below keeps getting garbled by board engine, see original website!
ruby -e "$(curl -fsSL )”
3. Setup PATH:
echo “export PATH=$PATH:~/abin” >> .bash_profile
4. Install
by
Peter Molfese
-
AFNI Message Board
See this thread about one way to get around the new System Integrity Protection (SIP) "feature" in Mac OS 10.11.
I've found a few other work arounds, but have only tested them on one computer at this point.
by
Peter Molfese
-
AFNI Message Board
Hi Ajay-
I believe that Rick and I were under the impression that you were doing functional MRI with SUMA. Based on your goals of doing anatomical analysis with both 1) cortical thickness and 2) subcortical structures, I would recommend sticking to a more Freesurfer based pipeline. That way there's considerably more documentation. If you're still interested in mapping fMRI data, I
by
Peter Molfese
-
AFNI Message Board
Hi Ajay-
As Rick mentioned, the use of 3dQwarp is unnecessary here, because there is a better way. That better way is using standard surfaces. Essentially you're identifying common nodes among all of your Freesurfer surfaces. See Page 55 of this document.
QuoteOn such standard meshes, functional data mapped to node n on one subject can be directly compared to data mapped to node n
by
Peter Molfese
-
AFNI Message Board
To add on to Rick's answer. SUMA_AlignToExperiment uses a similar approach to Freesurfer's FsFast, as they also rely on a affine transform to work with either FSL or SPM. More information here:
surfer.nmr.mgh.harvard.edu/fswiki/FsFastRegisterToStructural
by
Peter Molfese
-
AFNI Message Board
Ok, let's start with the hard to imagine ones:
1) Try updating your binaries. Worth trying.
2) Do you have enough hard drive space? what's the output of "df -h"?
3) What's the output of "ls -al" on the directory with your data?
4) What's the output of "csrutil status"?
5) Have you tried to chmod 775 the directory?
6) Try fixing permis
by
Peter Molfese
-
AFNI Message Board
Also, just to be sure... what's the output of:
afni_system_check.py -check_all
We're running 10.11 on a number of systems here and haven't run into this issue.
by
Peter Molfese
-
AFNI Message Board
There are two "popular" pipelines for using surfaces. The first, which is what I think you're talking about, is to process your data in the volume (afni_proc.py recommended) and then visualize the results on the surfaces. This makes use of traditional processing techniques like smoothing in the volume, and can use 3dQwarp for nonlinear registration.
The second approach is to
by
Peter Molfese
-
AFNI Message Board
Hello,
Yes the most straightforward way of registering to a different template and then projecting the results onto that template in SUMA it is to use Freesurfer to create surfaces. AFNI does not currently have a method for creating cortical surfaces.
-Peter
by
Peter Molfese
-
AFNI Message Board
Hi Paul,
Can you post your layout file? You should be able to specify the controller location as well, from the README.environment:
***LAYOUT
A geom=+73+1106 // start controller A
A.sagittalimage geom=320x320+59+159 ifrac=0.8 // and Sagittal image
A.sagittalgraph geom=570x440+490+147 matrix=9 // and Sagittal graph
B
by
Peter Molfese
-
AFNI Message Board
I only had that happen once, one thing to try is using the AFNI layout files. Once you retrieve the files using Mission Control, use AFNI to: Define Datamode --> Misc --> Save Layout. Then you can play with the starting places of each window/controller. You can use a layout as the default, or launch AFNI with -layout <some_file>.
With our multi-monitor setups, we then alias AF
by
Peter Molfese
-
AFNI Message Board
Once you have labels created in Freesurfer, you just need to convert them to GIFTI format (or NIFTI if it's a volume). So if I wanted to take BA45 from the fsaverage to an individual subject and convert it to GIFTI (.gii), I could do something like this:
mri_label2label --srcsubject fsaverage \
--srclabel fsaverage/label/lh.BA45.label \
--trgsubject Subject1 \
--trglabel Subject1/la
by
Peter Molfese
-
AFNI Message Board
Dan-
How did you make your surface ROIs? If they're coming from Freesurfer Labels, you might be able to use mris_label_calc out of the Freesurfer distribution. If you're comfortable with MATLAB, there may be some ways to do this in the Surfing toolbox, though I've not tried yet.
- Pete
by
Peter Molfese
-
AFNI Message Board
We structured the Pediatric Atlas to be like an on-ramp, so the ages in the Pediatric Atlas are normally distributed in age with the mean of 10. I suspect the deformation distance (amount of warping you have to do) would be smaller using our Atlas than using one of the adult atlases.
I would really like you to use the Haskins Pediatric Atlas, but you could also consider one of the newer MNI
by
Peter Molfese
-
AFNI Message Board
Hmm, I have SUMA up and running at the moment on 10.11, perhaps try reinstalling XQuartz?
This release actually marked a shift for me, I waited almost a week to install it!
by
Peter Molfese
-
AFNI Message Board
Just a warning for users upgrading to Mac OS X 10.11 "El Captain". A new security feature called "System Integrity Protection" prevents Python from directly accessing variables such as DYLD_FALLBACK_LIBRARY_PATH. You may notice that the Python superscripts (e.g. afni_proc.py, align_epi_anat.py) may fail at 3dSkullStrip or 3dAllineate. While running these programs individual
by
Peter Molfese
-
AFNI Message Board
To add onto Daniel and Zhark's comments. It looks like you are trying to estimate the smoothness of the data from the actual ANOVA results. This is... not ideal. Instead you should use 3dFWHMx to estimate the smoothness of your individual subjects errts files, calculate the average of this smoothness value, and use those values to put into your 3dClustSim command. Relevant warning from 3
by
Peter Molfese
-
AFNI Message Board
Hi Natalie-
Have you tried running the command without a mask? Also, when you install the R packages (afex, phia, snow), do you get any error messages? There is a little bit of a lag in some of the dependencies for the "snow" package that have been causing some issues.
What version of R are you running? And are you on a Mac?
-Peter
by
Peter Molfese
-
AFNI Message Board
So you either need to put all of the commands in a single line. Or add a \ at the end of each line, while also removing empty lines entirely.
by
Peter Molfese
-
AFNI Message Board