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
Results 3961 - 3990 of 4524
The FDR curves in the stats dataset output by the afni_proc.py
script are added automatically by 3dDeconvolve. The script is
not running 3dFDR.
I would expect the default to be -cind, but could not tell you when
-cdep might be more appropriate.
- rick
by
rick reynolds
-
AFNI Message Board
Hi ASI,
1. What do you mean it is not precise enough? Why is the
correlation value not what you want?
Are you complaining that the result is not stored in a
dataset that you can access from the command line? If
so, just save the overlay (Define Datamode -> SaveAs -> OLay).
2. That is right. Having processed the data with afni_proc.py,
your input to InstaCorr w
by
rick reynolds
-
AFNI Message Board
This error implies that the Condition1/2.txt files are being
viewed as input EPI datasets, meaning they are not preceded by
any -regress_stim_times option.
As Nick said, it would be helpful to for you to post the actual
command along with the error.
Check on how/if you used the -regress_stim_times option.
- rick
by
rick reynolds
-
AFNI Message Board
uber_subject.py does not scrutinize the datasets to
guess what might have been done already, it is meant
to do fairly standard processing.
Also, you should pass your motion parameters to
afni_proc.py to use in censoring and/or regression.
See -regress_motion_file in the -help output.
Please review the afni_proc.py -help output, through
example #6. But basically, you will just do what
by
rick reynolds
-
AFNI Message Board
If the datasets have already been put into MNI space,
that is significant pre-processing. Then you should not
be telling afni_proc.py to do any of: tshift, align, tlrc or
volreg.
If you want to use the command you gave, the data should
be unprocessed.
- rick
by
rick reynolds
-
AFNI Message Board
Yeah, that code needs to be fixed. WARNING_message()
was basically designed to take numbers as formatting
arguments, but in this case it is taking a huge string. And
using a va_list, it does not even see those arguments to
check the size on. Hmmmm...
- rick
by
rick reynolds
-
AFNI Message Board
Hi Paul,
That is probably not related to any change in AFNI (no one is
touching that code right now, I believe). But you can turn it
off by setting AFNI_VERSION_CHECK to No. Otherwise,
such a check happens every 5 days (assuming the check
succeeds).
Send me email with your IP address, so that I can make sure
you weren't accidentally blacklisted on our end.
- rick
by
rick reynolds
-
AFNI Message Board
It would help if you provided more details here. The 10.7
package should work fine if that variable is set. This should
have nothing to do with XQuartz.
If this is not working, please give the complete output of the
following commands:
which afni
afni -ver
echo fallback = $DYLD_FALLBACK_LIBRARY_PATH
echo $0
- rick
by
rick reynolds
-
AFNI Message Board
Hi Nick,
So the output of 3dTshift has sform_code=1, and from
3dAllineate it is 4? If so, would you be willing to upload
the exact command and data necessary to repeat the
3dAllienate step?
Thanks,
- rick
by
rick reynolds
-
AFNI Message Board
Hi Zhiyong,
That sounds okay, though just report what you are doing. Note
that you would be starting with a significant cluster and then
restricting it to an anatomical ROI. That should be okay.
Consider also 'whereami -omask MASK ...', where MASK is the
ordered cluster mask from your clusters (i.e. from the 'SaveMsk'
button in the Clusterize Report, or from 3dmerg
by
rick reynolds
-
AFNI Message Board
You don't say what distribution and version you
are using, so it is hard to answer.
What is the output of these commands:
afni -ver
which R
- rick
by
rick reynolds
-
AFNI Message Board
Hi Thomas,
Just to add a little to this, note that the case of Siemens mosaics
is where we actually do find the slice timing. Yes, the within-slice
orientation is not really coded well in the DICOM file, but that does
not matter much for most processing. The key is that the location of
the data in space is correct, even if the scanner is reorienting each
slice.
So your slice directio
by
rick reynolds
-
AFNI Message Board
Hi Isaac,
Basically, your dataset mask is so small that cluster simulation is
not stable. You are getting tons of warning messages which are
being accumulated before being printed out (so that they appear
at the end, I expect, so that one actually sees them).
In the case of -LOTS (and -niml implies it, apparently), there are
many pthr values to simulate at, and each one generates many
by
rick reynolds
-
AFNI Message Board
Hi Nick,
I cannot quite reproduce this, it works fine from the beginning.
The temporary dataset that I see is in orig space, while your
error (and rm command) shows tlrc space. That is peculiar.
Maybe your NIfTI dataset has sform_code=2, meaning it may
be viewed as in tlrc space to begin with.
What is the output of this:
nifti_tool -disp_hdr -field sform_code -infile m01_fwhm0
by
rick reynolds
-
AFNI Message Board
Volumes that are shown to be in alignment have presumably
been processed (by some other piece of software). Did you
read the link that I sent about this?
AFNI message board post
Of course this begs the question of exactly what registration
was done to the data. You should review where your datasets
came from.
- rick
by
rick reynolds
-
AFNI Message Board
Hi Zhiyong,
If it is simply that the cluster is too big and contains voxels outside
the amygdala, that is okay. Just start with a fully clustered volume
and then multiply by an amygdala mask.
However, if the clustering approach is simply missing the amygdala,
that is more difficult to deal with. Is this the case for you?
- rick
by
rick reynolds
-
AFNI Message Board
Hi Piero,
While my intention was to have uber_proc.py running full analyses
by now, I have not really gotten it started yet.
For now (and the next couple of years, probably), stick to using
uber_subject.py directly for single subject analysis and uber_ttest.py
for group t-tests (or using 3dMEMA).
I will add something to say not to use that program yet.
- rick
by
rick reynolds
-
AFNI Message Board
You need to fix any errors. There will not be a proc file
unless afni_proc.py runs without errors.
The error you are reporting is from the right button, when
you run the resulting proc script. But if there is no proc
script, then you need to look at the output from the actual
afni_proc.py command, which is reported when hitting the
middle action button.
Basically, there are 3 buttons
by
rick reynolds
-
AFNI Message Board
Hi Dante,
Does that mean the DYLD_FALLBACK_LIBRARY_PATH solution is
still not working, even with the 10.7 package?
If you use the 10.7 package, the install instructions are
basically:
0. install Xcode
1. run @update.afni.binaries -package macosx_10.7_Intel_64
(or download and install that package somewhere)
2. update PATH and DYLD_FALLBACK_LIBRARY_PATH variables
- ric
by
rick reynolds
-
AFNI Message Board
Exactly what processing has already been done?
And does that sform_code of 2 mean that the EPIs are
aligned to the anatomy, or are they already in some
standard space? Or are they simply aligned to an EPI
base?
It might be better to start with unprocessed EPI data,
but if you want to stick with some of the preprocessing
in another package, then just tell afni_proc.py not to
add those
by
rick reynolds
-
AFNI Message Board
If you are referring to the 10.7 binaries, that system had
been unavailable for more than a month now.
However it is now back in service and the binaries are
ready, yay. :)
If it is not 10.7 that you are after, please let me know.
- rick
by
rick reynolds
-
AFNI Message Board
Hi Steph,
I think the next one will be in September sometime. We may
already have dates for it.
The February class is already full, unfortunately. There is
space to sit in, but not for a desk with power for a laptop.
- rick
by
rick reynolds
-
AFNI Message Board
Hi Dante,
I just noticed that you said you followed the 10.6 instructions.
Those libraries are not in the 10.6 package, they are in the 10.7
package. Please use that instead. You can update to 10.7 via
the command:
@update.afni.binaries -package macosx_10.7_Intel_64
- rick
by
rick reynolds
-
AFNI Message Board
Hi Colm,
That is correct, the 10.7 binaries have not been updated
since Dec 20. I will have to take a peek and see why.
- rick
by
rick reynolds
-
AFNI Message Board
If you just type it on the command line, then do not log
out and back in. I was meaning that the line should go
in your .cshrc file.
Again, if you want to try it on the command line, type
it in then immediately try "3dSkullStrip -help", for example.
- rick
by
rick reynolds
-
AFNI Message Board
Hi Dante,
The export command is for bash users. If you are using tcsh, try:
setenv DYLD_FALLBACK_LIBRARY_PATH $HOME/abin
- rick
by
rick reynolds
-
AFNI Message Board
The output from afni_proc.py (which hopefully appeared on
your screen) shows that there are no stimulus labels.
Is the 'label' column of the stimulus table empty?
Maybe I should have uber_subject.py report that as an error,
rather than relying on afni_proc.py to report it. Not sure...
- rick
by
rick reynolds
-
AFNI Message Board
It would be helpful if you could tell me more than this.
If you get that again, consider (immediately) packaging
the directory for that subject (subject_results/group.GROUP/subj.SUBJ)
and emailing it to me as a tgz file.
Please send me a PM if you wish to do so.
- rick
by
rick reynolds
-
AFNI Message Board
Hi Pinson,
Keep in mind that SPMG1 is a fixed-shape model, meaning that
in using it you are ASSUMING it is the shape of the BOLD HRF
(similar to BLOCK or GAM). Any iresp curves generated from
that are fairly uninformative, because they will all look
like the SPMG1 curve, and will only vary in their magnitudes
(the magnitude is the beta weight, or Coef value).
TENT(0,16,9) can look lik
by
rick reynolds
-
AFNI Message Board
This is too hard to guess at. Please provide the resulting
afni_proc.py command, perhaps.
Update your binaries as Peter stated and see how it goes.
- rick
by
rick reynolds
-
AFNI Message Board