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 4471 - 4500 of 4524
Hi Mingbo,
Please see AFNI_NO_OBLIQUE_WARNING in README.environment
(should be in your abin). The README suggests it is only for the GUI
though, but I believe it applies to all programs.
- rick
by
rick reynolds
-
AFNI Message Board
Meaning the download failed? If you would like to be mailed a copy,
please use "Private Message" in the upper right to send a message
to the site administrator.
- rick
by
rick reynolds
-
AFNI Message Board
Hi Jatin,
The fitts dataset is created in the 3dTfitter command on that web site, but that
is because 3dTfitter does not have an -errts option. HJ shows usage of 3dcalc
to get the errts dataset (which he names as 'cleaned').
The major components of global WM and CSF seem to come from GM. That
is why it is not recommended.
Yes, HJ finds local WM signals (within some neighbo
by
rick reynolds
-
AFNI Message Board
Hi David,
1. Since 3.5*sqrt(3) would truncate, round up when applying it. You do not
need to round to an integer though (and 7 is 2 voxels). You could use 6.07.
2. The 3dClustsim threshold is in voxels, so multiply by 6.07.
3. That looks okay from the help, but I have not used it.
Note that you can use the Clusterize part of the afni GUI to do this, and
even have it print out a corre
by
rick reynolds
-
AFNI Message Board
Hi Jatin,
Actually, the errts_Decon+tlrc dataset is what you would start with
for a correlation analysis. It is the remaining signal after removing
the known sources of noise.
Note that in general, it is not great to remove any global signals,
even including white matter and CSF. It becomes a bit like global
signal regression. The strength of ANATICOR comes from a local
white matte
by
rick reynolds
-
AFNI Message Board
Hi David,
For use of -GOFORIT, apply that "directly" to 3dDeconvolve by using afni_proc.py's
-regress_opts_3dD option. Parameters to that AP option are passed directly to
3dDeconvolvle, e.g.
-regress_opts_3dD -jobs 6 -GOFORIT 9 -gltsym 'SYM: +houses -faces'
For the second question, not currently. That has been a personal item (meaning I
want it) on my tod
by
rick reynolds
-
AFNI Message Board
For that there is a 3dsvm program, written by Stephen LaConte, as well as a
plugin to the main afni GUI.
For installation, note that AFNI does not run on Windows (mac or linux are
better choices, if possible). With Windows, that gives you two main choices:
1. probably the easiest: install a virtual Linux OS using Virtual Box, or similar
2. harder, but better processing: install Linux a
by
rick reynolds
-
AFNI Message Board
Hi Mingbo,
Try "3dinfo -nv DSET", where -nv stands for nvals (or -nt).
See "3dinfo -help" for more details.
- rick
by
rick reynolds
-
AFNI Message Board
It might be better for you to describe details of exactly
what you want to use AFNI for, including study type and
specific software functionality that you are looking for.
- rick
by
rick reynolds
-
AFNI Message Board
Once I get a 10.7 machine to try this on, I will get back to you.
There were no problems running such a command on my linux
system.
- rick
by
rick reynolds
-
AFNI Message Board
You don't get to control that aspect with 3dvolreg. I suggest
sticking with using 3dTshift directly.
Actually, I suggest using afni_proc.py for the single subject
processing.
- rick
by
rick reynolds
-
AFNI Message Board
Hi Tony,
Include the output directory in the -prefix, as in:
3dTshift -prefix ../../some/other/dir/run1_shifted ...
- rick
by
rick reynolds
-
AFNI Message Board
For help, please describe your OS and version, along with
what you have done up to now.
Please see http://afni.nimh.nih.gov/pub/dist/HOWTO/howto/ht00_inst/html/index.shtml.
- rick
by
rick reynolds
-
AFNI Message Board
Hi Corianne,
If there is just one mask value you want to go after, you can restrict the -mask
input to it, in which case 3dROIstats should just output the related stats. Note
that in general, when AFNI programs read in a dataset, the values can be restricted
in a way that is similar to sub-brick selection.
To exemplify, if a dataset has many integral values, but only values of 17 are ca
by
rick reynolds
-
AFNI Message Board
The default polort in an afni_proc.py script is based on the duration of your
time series, rather than being a fixed degree.
What help are you referring to? Note that afni_proc.py can do all of
the pre-processing, including regression with censoring and bandpassing,
along with motion and its derivative. See example #9 from the help.
- rick
by
rick reynolds
-
AFNI Message Board
Yes, there are many correlation programs. For this particular use,
take a look at 3dTcorrelate.
- rick
by
rick reynolds
-
AFNI Message Board
Hi HP,
I am inclined to think 2, depending on how the bandpassing is done.
Note that 3dBandpass uses quadratic detrending by default. So if
that has already been done, polort 0 should be enough in 3dDeconvolve.
Note also that afni_proc.py does not do anything special with the polort
value when bandpassing in 3dDeconvolve.
- rick
by
rick reynolds
-
AFNI Message Board
The gain of such syntax is so that we don't have to modify 200 programs
to take a new option. With such a syntax change, they all work the same
way immediately. And any new programs automatically allow it.
This syntax would be akin to the current "DSET DSET DSET ..." syntax.
Anyway, we will ponder.
- rick
by
rick reynolds
-
AFNI Message Board
Ah, that's a different beast that really isn't specific to 3dbucket.
Maybe we could ponder some syntax like filelist:MY_FILES.txt as a
way to have programs read MY_FILES.txt as a list of input datasets.
Will ponder a bit...
- rick
by
rick reynolds
-
AFNI Message Board
Hi Colm,
Just a quick comment about request 1). Consider the use of backward
quotes, as in:
3dbucket -fbuc -aglueto DSET `cat listOfFilesToPutInBucket.txt`
- rick
by
rick reynolds
-
AFNI Message Board
Hi Michael,
For better or worse, that is the default of how most programs work with volume
datasets. Try it with 3dcalc for example. To override that, use ./ in the output
dataset name, e.g. -cbucket ./001_yc_coefs.
- rick
by
rick reynolds
-
AFNI Message Board
Hi Andrew,
If you mean the warnings from 1d_tool.py in the afni_proc.py script,
the "IDENTICAL", "high" and "medium" warnings are just that, warnings.
It does not know whether they are problems, but you are warned to at
least think about it.
It is basically impossible for the program to "know" whether there is
a problem.
By default, afni_procp.
by
rick reynolds
-
AFNI Message Board
Hi Karin,
1. The pre-convolved functions are just boxcars at the stim onset
times of the duration specify in the basis function. There is
little to see.
If you really want to see, convert the stim_times to 1D files
using timing_tool.py, following example 6 from the help. This
will not model amplitudes, but that is just your list of them
with the mean removed.
by
rick reynolds
-
AFNI Message Board
Hi Kyle,
I expect that you are not running /sw/bin/python but are
running /usr/bin/python (check with "which python" ). If so,
the help instructs you to put /sw/bin before /usr/bin in your
path (in .cshrc).
Once that is done, you should probably logout and log back in,
to clear your path.
- rick
by
rick reynolds
-
AFNI Message Board
That's not needed! I'm just offering my opinions on this... :)
- rick
by
rick reynolds
-
AFNI Message Board
I think it makes most sense to stick to the resolution of the data,
both for the EPI and the anat (so as not to over- or understate what
you acquired).
Since the antomies were aligned to the MNI template, the data is
considered to be in MNI space. Being between voxel centers seems
fine. The resolution of the anatomies should not matter much, they
are generally only used for alignment (
by
rick reynolds
-
AFNI Message Board
Ah right, the templates are 2 mm^3. Maybe either way makes sense.
The coordinates you will report will probably come from the EPI
resolution data in any case, which is at 3 mm^3. That means only
1/8 of the voxels land on the 2 mm^3 voxels anyway, which to me
seems to say that you might as well use 1 mm^3 for your anats,
since they do not really matter in all of this (unless you report
s
by
rick reynolds
-
AFNI Message Board
Hi Georgios,
Really, you could use either. If you are reporting results per subject,
then it would make sense to show the anat of each subject that you show
stats for (again, IF you show stats per subject). But otherwise, the
templates are more commonly used.
Note that it is usually not a good idea to count voxels, but rather to
report volumes. That way the voxel size does not matter
by
rick reynolds
-
AFNI Message Board
Note that -dxyz can be used for isotropic voxels.
For the first command, I am not sure that there is an reason
to undersample your subject anatomy to match the template.
But yes, those commands seem good.
---
In general, we hope you are already using afni_proc.py, and
in such a case -volreg_warp_dxyz is not actually needed, since
that is what it does by default (truncates smallest di
by
rick reynolds
-
AFNI Message Board