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 - 20 of 20
Hi Paul,
I'm logging in from my mac to the NCF computer cluster. Right-clicking on the Olay button to the left of the sub-brik selector doesn't do anything, but I can right-click on other locations (e.g., if I right-click the Olay button above the Clusterize button, a menu pops up). This is the case whether I use noVNC to log in or XQuartz. Left-clicking provides the large menu that
by
Laura
-
AFNI Message Board
Thanks so much for your quick response! I'm using precompiled binary linux openmp 64: Nov 19 2018 (Version AFNI 18.3.05). I'm able to right click other menus in the GUI, but for some reason when I right click the Olay button next to the sub-brik selector, nothing happens. I'm not sure if it's a noVNC or AFNI issue. Even so, when I left click, I don't have a scroll bar lik
by
Laura
-
AFNI Message Board
Hi all,
Sorry to post this topic given that it's been answered before, but the previous solutions haven't worked for me. I have 150 sub-briks or so in my overlay file, but I can only see 99 sub-briks in the sub-brik selector dropdown menu. When I expand the browser to full size, I can see a bit more but not the full number of sub-briks. I've tried right-clicking next to the OLAY
by
Laura
-
AFNI Message Board
Hi all,
Unfortunately my worst nightmare was realized, and even though it looks like the group x condition interaction was significant (see post above), the post hoc GLTs are all non-significant (at FDR-corrected 0.05). How is that possible? Am I coding the GLTs incorrectly, missing post-hoc tests, reading the interaction incorrectly (I'm setting the interaction sub-brick for both the ove
by
Laura
-
AFNI Message Board
Thanks, Paul! That was extremely helpful. I just needed to increase my allocated memory and then it ran! I did find a significant group by condition interaction when FDR corrected, so that makes me feel better, but now I have to learn to tackle the GLT coding and do some post-hoc digging. I'll be back if I can't figure it out!
Thanks,
Laura
by
Laura
-
AFNI Message Board
Hi Gang,
Thanks for your quick response! I'm running the analysis on our computer cluster at Harvard. All of the necessary software has to be loaded through a modules system every time we log on (included in my .bashrc-see below). When I run the afni_system_check.py -check_all, the necessary R packages seem to be recognized. Am I missing something? Should I try on my local machine to see
by
Laura
-
AFNI Message Board
Hi all,
I'm new to 3dMVM (used to run 3dANOVAX, but now my models are a bit too complex, and I have unequal sample sizes), and I'm having some trouble getting the program to run. First of all, I want to explore whether there is a significant group (3 groups: TD, RD, EL) by condition (8 conditions: SPEECH, NOISE, PSW, FF, OPplus, OPminus, SEM, UNREL) interaction. I'm worried that
by
Laura
-
AFNI Message Board
Hi all,
Sorry for this basic question, but I'm having a bit of trouble interpreting the output of the ss_review_basic script. It says that there are 58 TRs above the motion limit (which is not surprising given that we're scanning kids), and 9 above the outlier limit, but then the total number of censored TRs is 88. Why is this number so much higher than the combined number of TRs cen
by
Laura
-
AFNI Message Board
Hi Rick,
Sorry for all of the messages! Changing the polort still didn't work, so I'm wondering if you could let me know why using 3dSynthesize is wrong (in what ways) since that seems to work the best so far.
Thanks,
Laura
by
Laura
-
AFNI Message Board
Hi Rick,
Thanks so much for your suggestion! I've tried it to no avail, unfortunately. It's strange. When I enter the data from just before the regression step (e.g., pb04.$subj.task.r01.blur.nii) into the multivariate analysis software, it's able to pick up on all of the hypothesized networks in the different components that it outputs (after a few components clearly linked to
by
Laura
-
AFNI Message Board
Hi Rick,
Thanks so much for your response! I guess I'm just confused about which output to use for my multivariate analysis then since I'd ideally like one file that just has motion and WMe/Svent regressed out with no (or minimal) detrending (like an allruns file just with the motion and nuisance signals regressed out). It seems like no matter how lenient I'm being in the prepro
by
Laura
-
AFNI Message Board
Sorry to follow up with another question, Rick, but I've been assuming that I should be using the errts file as my output for the resting state analysis, but it still seems a bit too sparse for the multivariate analysis I'm trying to run. I'm wondering if I should actually be using 3dSynthesize to pull out the regressors of no interest and then subtracting that from my all_runs fil
by
Laura
-
AFNI Message Board
Hi Rick,
Thanks! Yes, that's what I want. I see where to change this in the script–Example 11b–for the CSFe in the ventricles (-regress_ROI SVent), but what about for the WMe? Is that just using a simple GLM-based approach or is that using a principle components approach in the background somewhere?
Best,
Laura
by
Laura
-
AFNI Message Board
Sorry to respond to this thread so many months later, but I was wondering how I can modify this script to do a simple GLM-based approach of removing WMe and CSFe (in the ventricles). The current script (regressing out the principle components) has been great for my univariate analyses using 3dGroupInCorr, but the resulting data are "too clean" for the multivariate analysis that I'm
by
Laura
-
AFNI Message Board
Hi Rick,
Thanks for your quick response! I'm running the analysis on a computing cluster, so I'm setting the memory limit. I've set it to 5,500 MB, but it looks like it only used about 3,638 MB on the last participant, so it might be okay. The datasets are about 512 MB large. I'm going to try using affine registration to see how much it speeds things up. If it's not si
by
Laura
-
AFNI Message Board
Hi Rick,
Thanks so much for all of your helpful responses! Just wanted to check in on a few final things (sorry). I only have one run, so I deleted the "-regress_ROI_PC_per_run" line. Just wanted to confirm that that's correct.
For the "-regress_ROI_RC" line, I used 3 as the number of principal components because that's the number used in the example. Is there
by
Laura
-
AFNI Message Board
Thanks so much for your response, Rick! Sorry to bother you again, but I just tried running it, and I keep getting the following error: ** error: option -regress_ROI_PC follows unknown arg #29 (-mask_import). I'm running the most current version of afni available on the cluster (December, 2015). Was the -mask_import option not available then or is there something strange going on with my scr
by
Laura
-
AFNI Message Board
Hi Rick,
Thanks for your quick reply! So if I decided to use Example 11b, I just want to be clear which options are necessary vs. those that are just suggestions. If I resample the template_ventricle mask to 2 mm isotropic instead of 2.5, I assume that I would I get rid of the -volreg_warp_dxyz 2.5 since the default output is 2 mm. I assume that I don't necessarily need to include "-
by
Laura
-
AFNI Message Board
Hi all,
I typically use afni_proc.py when analyzing resting state data (Example 9 when I don't have physio data, as is the case here), but I'm losing too many degrees of freedom due to band-pass filtering (e.g, losing 487 degrees of freedom out of 554 solely due to band-pass filtering). My resting state scan only consists of an approx. 6-minute-long run with a TR of 650 ms and a mult
by
Laura
-
AFNI Message Board