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
Page 3 of 5
Pages: 12345
Results 61 - 90 of 145
Another trick that I've found very helpful with noisy T1 images is to use @clip_volume to create a box (everything outside the box is set to 0) as small as possible around the brain thus manually removing as much of the noise and nonbrain tissue as possible. It has helped enormously when I was skull stripping noisy T1s as 3dSkullStrip had very little to cause it to expand the mask outside th
by
Colm Connolly
-
AFNI Message Board
I'd like to see censoring based on FD and DVARS integrated into afni_proc.py. With similar abilities to censor additional volumes to the left and right of volumes marked for censoring along with being able to parameterize FD and DVARS limits.
The ability to provide arguments like the following (from afni_rest_proc.py) to afni_proc.py would be great:
-censorleft 1
-censorright 2
-fdlim
by
Colm Connolly
-
AFNI Message Board
Hey Paul,
I wrote my own robust regression script which can also bootstrap the standard error. In either case it's parallelized to speed up the computations. The code is at
It is heavily tuned to my own needs but given the time I guess I could make it much more generic and more usable to others. Either way you're welcome to take the code and modify for you own needs should you de
by
Colm Connolly
-
AFNI Message Board
Hey Rick,
I just loaded the aparc+aseg and aparc.a2009s+aseg separately in freeview yesterday and verified the list of numbers in the example commands. In both cases the numbers referred to the same structures.
What caused me confusion yesterday was the mention of both atlases in the freesurfer note coupled with my experience of other atlases almost never overlapping in their choice of ROI
by
Colm Connolly
-
AFNI Message Board
Never mind, it seems that the ROI numbers work regardless of the atlas used. Sorry for the noise
by
Colm Connolly
-
AFNI Message Board
Hi,
I'm going through the examples from afni_proc.py to do resting state processing using afni_proc.py
The freesurfer note makes reference to aparc.a2009s+aseg but the example 3dcalc command then goes on to refer specifically to aparc+aseg.nii. My understanding of these different parcellations is that the come from different atlases. So my questions then is do the series of ROI numbers
by
Colm Connolly
-
AFNI Message Board
you may need to manually set the contents of TAXIS_OFFSETS in the header of the dataset for 3dTshift to perform time shifting.
it's been a while since I've had to do this but I think you can do it with 3drefit -atrfloat TAXIS_OFFSETS 'offsets' dset+orig
where 'offsets' (retain the quotes in the command) is replaced with the appropriate slice time offsets.
by
Colm Connolly
-
AFNI Message Board
The model formulae specification are standard R formulae. You can read here for some guidance
"A*B+c": main effect of a, main effect of B and their interaction with c as a covariate. This is equivalent to "A + B + A:B + c" where A:B indicates the interaction of A with B.
A:B+c models the interaction of A and B with c as a covariate
by
Colm Connolly
-
AFNI Message Board
I've just rerun this without the |SubjID in the random effects (it's just ~1) and i'm still getting the same large differences between the group F subbrik and the PAT-NCL Z.
There's no noticeable difference between with and without the |SubjID in the random effects formula.
image
by
Colm Connolly
-
AFNI Message Board
Gang Wrote:
-------------------------------------------------------
> Colm,
>
> A couple of comments:
>
> 1) The following line
>
> -ranEff '~1|Subj'
>
> should be
>
> -ranEff '~1'
>
> I'm not so sure if 3dLME would complain about the
> syntax.
It didn't complain at all.I'll try rerunning it
by
Colm Connolly
-
AFNI Message Board
Hi,
I'm having trouble with the output of 3dLME (I think)
I've been using it to investigate difference sin resting state connectivity and am running it as follows:
3dLME -prefix restingstate.patAndCtrl.group.and.gender.L_whole_amygdala.3mm.lme.bucket.20151102-1251PST
-jobs 8
-mask mask.grey.patAndCtrl.union.masked+tlrc.HEAD
-model 'Group*Gend
by
Colm Connolly
-
AFNI Message Board
Couldn't you use 3dbucket to extract only those sub-bricks you need for you need for your analysis and then feed those newly created buckets into 3dLME? This would seem to be one way around the problem you are describing. Granted it's wasteful of disk space, but the bucket file can be deleted when no longer needed and recreated from the original source file at will.
by
Colm Connolly
-
AFNI Message Board
Hey,
I made those seeds one at a time and saved them in separate files. After they are created, if you want them in one file you can simply use 3dcalc to add them together as they do not overlap
3dresample -dxyz 3 3 3 -inset $FSLDIR/data/standard/MNI152_T1_1mm.nii.gz -prefix ./MNI152_T1_3mm.nii
echo "-5 -10 37" | 3dUndump -master MNI152_T1_3mm_brain.nii.gz -srad 3 -orient LPI -p
by
Colm Connolly
-
AFNI Message Board
3dDeconvolve takes a -polort option where you can specify the degree of the polynomial you require. it defaults to 1 if not supplied.
if you are using afni_proc.py to perform your analysis the option you need it -regress_polort. As above, it takes an argument specifying the degree of the polynomial you want to use.
by
Colm Connolly
-
AFNI Message Board
Hi,
I wonder if it would be possible to remove the restriction that a covariates file cannot be provided to 3dttest++ when a using set (setA or setB) that has multiple sub-bricks from one dataset?
I have all of my sub-bricks labeled to match exactly the subject column of my covariates file. I envisage a use case where if all of the sub-brick labels can be matched exactly one-to-one with th
by
Colm Connolly
-
AFNI Message Board
Assuming you have real valued voxels in ${seed} that you want to use as weights in addition to their defining the mask, something like the following should do what you want.
3dcalc -datum float -a ${seed} -b input.nii.gz -expr "a*b" -prefix output
3dmaskave -quiet -mask ${seed} output+tlrc.HEAD > ${seedName}.ts.1D
However, a very useful feature to 3dmaskave would be the addit
by
Colm Connolly
-
AFNI Message Board
I don't use 3dLME myself as I generally code up my own, but I would have thought that you would need a + in your formulae:
-gltLabel 1 'HF-NF' -gltCode 1 'Task : 1*HiFat -1*NonFood ' \
-gltLabel 2 'LF-NF' -gltCode 2 'Task : 1*LowFat -1*NonFood ' \
-gltLabel 3 'HF-LF' -gltCode 3 'Task : 1*HiFat -1*LowFat ' \
-gltLabel 4
by
Colm Connolly
-
AFNI Message Board
Hey Danny,
3dTstat -mean will calculate the mean across all sub-briks and then you can use 3dcalc to perform the subtraction.
Bye,
Colm.
by
Colm Connolly
-
AFNI Message Board
yeah, I eventually figured out that auto_warp was the culprit. I've got a patch which almost completely works with compressed nii files. I've only got one issue left to fix. When it's done I can post it here if you want. It would, however, require a small patch to afni_proc.py since it expects uncompressed nii files. I've not written that latter one yet.
by
Colm Connolly
-
AFNI Message Board
Hi everybody,
I'm playing around with the afni_proc.py (from Precompiled binary linux_openmp_64: Feb 19 2015 (Version AFNI_2011_12_21_1014)) script for the first time.
I've run it as follows:
subj=330_AA
task=mid
cd ../data/$subj
afni_proc.py \
-subj_id $subj \
-script proc.$subj -scr_overwrite \
-do_block align tlrc \
-copy_anat $subj.anat+orig.HEA
by
Colm Connolly
-
AFNI Message Board
The following are a few good references on this topic
Costafreda SG, Brammer MJ, Veˆncio RZN, Moura ̃o ML, Portela LAP, et al. (2007) Multisite fMRI reproducibility of a motor task using identical MR systems. J Magn Reson Im 26: 1122–1126. doi:10.1002/jmri.21118.
Gountouna V-E, Job DE, McIntosh AM, Moorhead TWJ, Lymer GKL, et al. (2010) Functional Magnetic Resonance Imaging (fMRI) reproduci
by
Colm Connolly
-
AFNI Message Board
Hi everybody,
I wonder of any one can tell me if there is any way to provide non-Freesurfer derived GM and WM masks (in my case generated by FSL's FAST) to afni_restproc.py ?
From reading the help and code for afni_restproc.py the answer seems to be a no, which leaves me with two options
1. patch the afni_restproc.py code to accept separate masks for GM and WM masks and process the
by
Colm Connolly
-
AFNI Message Board
Hi,
I think I've discovered a bug in the way AFNI deals with clustering when two afni controller windows are open.
Version AFNI_2011_12_21_1014
[]
on MAC OS X 10.9.4
Here are the steps to repeat.
1. Start afni default underlay of MNI152_T1_1mm.nii.gz is automatically chosen
2. load the overlay
3. Select olay and thr to be the same subbrik.
4. Set the slider to some reasonab
by
Colm Connolly
-
AFNI Message Board
Applications accepted for the “CCC-UCSF Neuroscience of Creativity Postdoctoral Fellowship Award”
The University of California San Francisco (CA, USA) & Center for Childhood Creativity (CCC, CA USA) is awarding a two year postdoctoral fellowship to an individual with exceptional creativity in studying the neuroscience of how high level affective/motivational processing and learning interac
by
Colm Connolly
-
AFNI Message Board
Hi Justin,
You can use 3dAttribute to do this. The following bash function, which I use in my code, may be helpful to you:
function extractFStatpars {
local bucket=$1
local subbrikId=$2
a=$(3dAttribute BRICK_STATSYM $bucket\[$subbrikId\] )
b=${a##*(}
c=${b%%)*}
echo $( echo $c | tr "," " " )
}
Regards,
Colm.
by
Colm Connolly
-
AFNI Message Board
Please do not contact me about this position. Contact details are at the bottom of the advertisement:
Postdoc positions in Developmental Neuroscience at UCSF, San Francisco CA
***************************************
The Laboratory for Educational Neuroscience (brainLENS.org) at the University of California, San Francisco USA is seeking 1 postdoctoral fellow for a minimum commitment of two ye
by
Colm Connolly
-
AFNI Message Board
Thanks, Rick. I just thought it might have beens symptom of something more problematic.
by
Colm Connolly
-
AFNI Message Board
Hi,
Below is a cut'n'paste of an error (down at the bottom of this message) that I've been getting from 3dClustSim this morning on Mac OS X 10.9.1
The mask used in the invocation of the 3dClustSim command is at
> 3dClustSim -mask /Volumes/PROMISEPEGASUS/yangdata/cPine/data/rois/sgacc.bilateral.3mm+tlrc.HEAD -fwhm 4.2 -niml -prefix CStemp.fwhm4.2.sgacc.bilateral.3mm+tlr
by
Colm Connolly
-
AFNI Message Board
Since you are only interested in ROIs 1 and 2 you should probably just separate those out of your compound mask as using the data range restrictors may not necessarily do what you want.
To split out ROIs 1 and 2 you could something like
3dcalc -a mymask+tlrc. -expr "equals(a, 1) + 2 * equals(a, 2)" -prefix rois1and2only
by
Colm Connolly
-
AFNI Message Board
Gang Wrote:
-------------------------------------------------------
> You probably have already obtained the PPI effect
> for each group separately and that should help you
I'm unsure what you mean by this. Do you mean the average for the R in each of the ROIs identified by the clusterized LME results?
> interpret the group comparison. Other than that,
> I'm no
by
Colm Connolly
-
AFNI Message Board
Page 3 of 5
Pages: 12345