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
Postdoctoral position in Cognitive Neuroscience at Haskins Laboratories
Applications are solicited for a postdoctoral fellowship in Cognitive Neuroscience at Haskins Laboratories. Applicants should have a Ph.D. in psychology, linguistics, cognitive neuroscience, or a related field. The selected candidate will contribute to multiple funded NIH studies of speech perception, reading, word acquisi
by
Peter Molfese
-
AFNI Message Board
Try:
tr '\r' '\n' < inputFile.txt > outputFile.txt
Which just translates any \r's into \n's. If the problem is the combination r's and n's you may need to try:
tr '\r\n' '\n' < inputFile.txt > outputFile.txt
by
Peter Molfese
-
AFNI Message Board
Most (dare I say "all") AFNI programs can output NIFTI files if you just tack on the extension (.nii or .nii.gz). You can convert the entire stats file via 3dcopy
3dcopy stats.subject001+tlrc.HEAD stats.subject001.nii.gz
Or if you just want the t-statistic of a particular condition/contrast, you can use 3dbucket to pull out the bucket of interest:
3dbucket -prefix subject001_ts
by
Peter Molfese
-
AFNI Message Board
You can check the alignment of your anatomical (MPRAGE) and EPI by setting the underlay as the anatomical and the overlay as the EPI and setting the t-stat bar to 0.00. Alternatively, you can use @AddEdge. Alignment images can be saved using either the save button in the viewer or by constructing a script that uses the plugout drive in AFNI. @AddEdge can be very helpful and you can run it auto
by
Peter Molfese
-
AFNI Message Board
Fink isn't necessary, you can do it with your own installation of Python and PyQT. The easiest way that I have found is to use homebrew. Assuming you have Xcode installed, you can do the following:
brew install pyqt
export PYTHONPATH=/usr/local/lib/python2.7/site-packages:$PYTHONPATH
uber_subject.py
An alternative approach to using homebrew: I also got uber_subject.py working via
by
Peter Molfese
-
AFNI Message Board
Job Announcement – March 2013
Research Associate
Haskins Laboratories, 300 George Street, Suite 900, New Haven CT 06511
Project Directors: Dr. Kenneth Pugh, Dr. Jay Rueckl
Haskins Laboratories is seeking a Research Associate for projects related to Cognitive Neuroscience and Language/Reading Development as part of several recently funded NIH R01 and P01 grants.
Responsibilities will
by
Peter Molfese
-
AFNI Message Board
I had encountered this bug with 3dMVM a while ago, have you tried updating your binaries?
PM
by
Peter Molfese
-
AFNI Message Board
I would use a for loop in the shell:
for aNum in {1..100}
do
3dTcat -prefix subj001_${aNum}.nii "input_file+orig.HEAD[$aNum]"
done
You could get more fancy by using 3dinfo -nt to get the total number of time points.
by
Peter Molfese
-
AFNI Message Board
You should be able to use 3dRegAna to do this by specifying a model like:
-model 2:0 1
I found this link informative. Also see example 3 in the RegAna article (http://afni.nimh.nih.gov/pub/dist/doc/manual/3dRegAnam.pdf).
by
Peter Molfese
-
AFNI Message Board
Postdoctoral position in Cognitive Neuroscience and Reading Disability
Haskins Laboratories (PI: Dr. Julie A. Van Dyke), New Haven, CT Application deadline: April 15, 2013
Applications are solicited for a postdoctoral fellowship in Cognitive Neuroscience at Haskins Laboratories. Applicants should have a Ph.D. in psychology, linguistics, cognitive neuroscience, or a related field. The select
by
Peter Molfese
-
AFNI Message Board
Assuming that you only want to look at the cerebellum:
You could create a mask, either by hand or using one of the many atlases included in AFNI, see this for more info (http://afni.nimh.nih.gov/pub/dist/edu/latest/afni11_roi/). You could then use this as a mask for your group-level analyses (3dttest++, etc).
If you go the atlas route, you could do something like this:
whereami -mask
by
Peter Molfese
-
AFNI Message Board
If you want all of the voxels in a mask, I would recommend 3dmaskdump. For the average of the voxels in the mask you could use either 3dmaskave or 3dROIstats.
by
Peter Molfese
-
AFNI Message Board
As a note, if using neurodebian - it can be easier to use the following command:
sudo apt-get upgrade afni
by
Peter Molfese
-
AFNI Message Board
This can be accomplished a few ways. My preferred way is to use the interactive clustering method. The report function of interactive clustering will show you the clusters of a given size (or larger) for any t-value (and related p-value) that you wish. If you then wish to show each cluster as a different color, use the SaveMsk button to save your clusters as a mask. Each cluster will receive
by
Peter Molfese
-
AFNI Message Board
Hi Rick,
Baseline was poorly named in my script - it is a condition and separate from the actual "baseline". I will try the scaling step as well.
Peter
by
Peter Molfese
-
AFNI Message Board
Hi Rick-
0. The task is block design and has Condition A, Condition B, and rest. The map shown was A-Rest (it is in other conditions as well).
1. I ran 3dREMLfit in afni_proc.py via regress_reml_exec.
2. I'm not currently using the scaling block.
3. There isn't anything special defined as what goes into the baseline. The baseline on this particular subject is a little shor
by
Peter Molfese
-
AFNI Message Board
I'm having some interesting behavior with a new paradigm and differences between 3dDeconvolve and 3dREMLfit. 3dREMLfit seems to be finding a significant outline around the brain - roughly where the EPI coverage is. The outline is not present using 3dDeconvolve.
3dDeconvolve
3dREMLfit
I'm using a pretty standard afni_proc.py analysis. Any thoughts why this might be showing
by
Peter Molfese
-
AFNI Message Board
The error is usually related to your ROI mask and the functional dataset being different sizes. You can check this with 3dinfo -prefix -n4 -d3 func+tlrc roi_TT+tlrc. This should show that the two datasets differ, with usually the TT_N27 based ROI at a higher resolution than your functional dataset.
To fix, you will need to either run 3dfractionize or 3dresample. 3dfractionize has the adva
by
Peter Molfese
-
AFNI Message Board
Please tell us the following:
1. what do you get when you type "echo $SHELL" and press enter?
2. paste the contents of your .cshrc file.
by
Peter Molfese
-
AFNI Message Board
You will need to close the terminal window and open a new one. If it doesn't work, can you type "less .cshrc" and paste the results here?
by
Peter Molfese
-
AFNI Message Board
If you are using bash add the following to your .bashrc and .profile
export DYLD_FALLBACK_LIBRARY_PATH=$DYLD_FALLBACK_LIBRARY_PATH:~/abin
If you are using tcsh - from this thread:
nano .cshrc
add the text:
set path = ( $path ~/abin )
if ( $?DYLD_FALLBACK_LIBRARY_PATH ) then
setenv DYLD_FALLBACK_LIBRARY_PATH $DYLD_FALLBACK_LIBRARY_PATH:~/abin
else
setenv DYLD_FALLBACK_LIBRA
by
Peter Molfese
-
AFNI Message Board
It looks like you don't have Time specified as a within subject variable (-wsVars). I've done the same thing before.
by
Peter Molfese
-
AFNI Message Board
AFNI Gurus-
I'm working with data from another site and have a couple of questions regarding how DIMON handles slice timing information.
On some of the newer collected data, output from Dimon shows that it is using a Siemens timing of alt+z2. The log sheet that was filled out when the data was collected says the data was collected in Altplus (alt+z). Is Dimon correct? Dimon output
by
Peter Molfese
-
AFNI Message Board
Hi Rick,
I have had some success bundling my python scripts using Freeze (linux) or py2app (Mac) to make them standalone executables including all of the dependencies. That may work for the PyQt4 programs in AFNI.
Peter
by
Peter Molfese
-
AFNI Message Board
I believe the macosx_10.7_Intel_64 binaries are "no glib" variety. I'm running them on 10.8 without fink or macports. I just add AFNI to the path and add the AFNI folder to the DYLD_FALLBACK_LIBRARY_PATH.
Are there specific programs that aren't working? Like 3dSkullStrip? Or is everything currently inoperable?
by
Peter Molfese
-
AFNI Message Board
Perhaps just a background color change. I really appreciate seeing the TENT functions in voxels that aren't activated - it helps me see the boundaries between activating above a statistical threshold and those not activating above threshold.
by
Peter Molfese
-
AFNI Message Board
Rick,
Is there any chance that some of those tcsh commands are not using a "tcsh -f" to skip over the global loading?
Peter
by
Peter Molfese
-
AFNI Message Board
What is the output of afni --version and can you post the entire output of the process?
by
Peter Molfese
-
AFNI Message Board