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 1 of 1 Pages: 1
Results 1 - 19 of 19
Hi Peter,
I see. Great!
And, thank you for all your help Peter!
Best,
Marina
by
Marina
-
AFNI Message Board
Hi Rick,
Sorry for the delayed response! This makes sense to me now. Thanks a lot for the explanation!
Best,
Marina
by
Marina
-
AFNI Message Board
Hi Peter,
I ran 3dmerge with the parameters given by the 3dclust or savemask, without using the -dxyz=1, and it worked. It did give me the same two clusters I saw on afni gui.
Thanks a lot for your help. I still would like to know why 3dclust and savemask reads the resolution of my file incorrectly (as I mentioned in my last email). There was some mistake in what I copied in the l
by
Marina
-
AFNI Message Board
Hi Peter,
Thank you for your response. I tried both the 3dclust and savemask buttons on the gui, and below are what I got. The common problem with both outputs are that -dxyz=1, indicating that it is reading the file as having a resolution of 1x1x1. Our file has a resolution of 3x3x3 instead. How can I have the the file resolution be recognized correctly ? Thanks!
Best,
Marina
++ 3d
by
Marina
-
AFNI Message Board
Hi Peter,
Ok, the parameters I used for clusterize option on afni gui is:
p =.0099, t=2.924, NN level=2, Voxel = 33,
input file: Conjunctiontemp_CR12FA56_and_CR12HiConfiProportionalHit_17Subjects_3x3_2only_ExclusivelyMasked_wTvaluetemp_Wholebrain+tlrc
The commands I used for 3dmerge is below. Note that our resolution is 3x3x3, so the voxel size is 33x(3x3x3)=891 mm3. Th
by
Marina
-
AFNI Message Board
Hi,
I am seeing different numbers of clusters when using 3dmerge to clust_order an analysis as opposed to using the clusterize option on afni gui. I think I used the same parameters for both methods (source subbrick, threshold, rmm, etc). It used to work and the two ways came up with the same results, but not today. Is there any idea about what could potentially be wrong here?
Best,
M
by
Marina
-
AFNI Message Board
Hi Rick,
We don't have a file that is named output.proc. We have files in such names:
out.cormat_warn.txt
out.pre_ss_warn.txt
out.ss_review.SS01.txt
Does any of those sound close to what you want?
Marina
by
Marina
-
AFNI Message Board
Hi Rick,
A correction, although the same error message came up, this time actually the script did not stopped running. It went on and finished!
How should we treat the error message in this case?
Best,
Marina
by
Marina
-
AFNI Message Board
Hi Rick,
Yes, the script stopped running there.
I now put in '-anat_has_skull no' and ran it again. The exact same problem came up.
Best,
Marina
by
Marina
-
AFNI Message Board
Hi,
We are making progress with afni_proc.py (with the updated version of afni). We ran into one problem though, and got this error message: THD_generic_detrend_LSQ: fit fails - no detrending.
Below are the commands we use (we are running it for all runs for Subject 1 for now):
afni_proc.py -subj_id afni_proc_Results \
-blocks despike tshift align volreg regress \
-copy_anat ana
by
Marina
-
AFNI Message Board
Hi Ziad,
That is correct. That comment was only added for this post. It isn't in the real code that I ran. Thanks for the advice.
Best,
Marina
by
Marina
-
AFNI Message Board
Hi Rick,
Thanks for your reply. Now I am using afni_proc.py following example 11, as suggested. We get this error message and can't figure out what is causing the problem here.
** error: option -tshift_opts_ts follows unknown arg #11 (-anat_follower_ROI)
----------------------------------------------------------------------
** failed command (get_user_opts):
Below is
by
Marina
-
AFNI Message Board
Hi,
I can't seem to find it myself among AFNI programs and thus am asking here if there is an AFNI program that does CompCor, i.e., a program that reduces noise in BOLD data with a component based method (as disucssed in Behazadi et al., 2007, titled 'A component based noise correction method (CompCor) for BOLD and perfusion based fMRI).
An additional question to ask, if t
by
Marina
-
AFNI Message Board
I would like to clarify on some details regarding my question. For the example I gave, let's assume that each voxel in the positive cluster has a T value of 3, so that every voxel survived the thresholding. Similarly, let's assume that each voxel in the negative cluster has a T value of -3, so that every voxel also survived the thresholding. Then the question is what I put in the last m
by
Marina
-
AFNI Message Board
Hi
I have a question regarding the command of clust_order and the direction of activation. It might be easier to start with an example. For example, we have two areas that survived a regression analysis, one is doing positive regression and the other is doing negative regression. The two areas happened to border each other. Let's say the positive area has a T value of 3, and has 51 vo
by
Marina
-
AFNI Message Board
Hi Rick,
Please disregard my most recent reply! Now we figured out what the problem is here. In our design, the 3.0 in our 3dmerge command should correspond to NN 2 on AFNI, instead of 3. And that is why there was a mismatch when I did NN 3. There was never a problem before because in what I have done, probably because there were no clusters of voxels that touched at corners and thus NN 2 an
by
Marina
-
AFNI Message Board
Hi Rick,
The problem here is when I use NN = 3, the cluster table and the GUI do not match. And usually they do, and they should. For example, when I use NN = 2, the cluster table and the GUI match. Similarly, on other analyses, the cluster table and the GUI match do.
Best,
Marina
by
Marina
-
AFNI Message Board
Hi Rick,
Thank you for your reply. I happened to try changing the NN level before I saw your reply. It turned out that if NN level is 2, the table comes with 17 clusters, and each corresponds with the clusters on GUI.
I tried to match the clust table of NN level 2 with that of NN level 3 (which has 16 clusters). It turns out that the NN_3 table has a cluster (cluster No. 3 on the li
by
Marina
-
AFNI Message Board
I conducted an ANOVA between three conditions and masked out the medial temporal lobe so that the rest of the voxels remained in the mask. Afterward, I did cluster_ordered the clusters. This is the cluster_order command I used: 3dmerge -1clust_order 3.0 384 -1tindex 1 -1thresh 5.46 -prefix ANOVA_HitFACR_Wholebrain_clust01 ANOVA_HitFACR_Wholebrain+tlrc.
The problem I have is that on the brain
by
Marina
-
AFNI Message Board