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 211 - 240 of 2632
Carolin, try something like
3dDeconvolve -stim_file your-regressor
In addition use '-polort 0' to include an intercept (or '-polort -1' to exclude an intercept).
by
gang
-
AFNI Message Board
Check your AFNI version:
afni -ver
If you don't have the most recent version, update your AFNI:
@update.afni.binaries -defaults -do_extras
Also, you don't have to add this line:
-glfLabel 1 sleepstage -glfCode 1 'sleepstage : 1*W -1*N2 & 1*W -1*N3 & 1*N2 -1*N3' \
3dMVM automatically provides the statistics for all the main effects and interactions.
by
gang
-
AFNI Message Board
Sarah,
3dLME should work fine with any size of mask. Make sure you have the most recent version of AFNI before your run your new 3dLME script. Alternatively, you could just apply the mask using 3dcalc to the 3dLME output from your original script (with the whole-brain mask).
by
gang
-
AFNI Message Board
> No individual fixed effect of diagnosis is included in the specified LMM-1 model but it yields z and p-values for the fixed effects?
> Two-way interactions are not specified in the model?
The R notation for the model formulation
Diagnosis *Age* Sex
is equivalent to
Diagnosis + Age + Sex + Diagnosis:Age + Diagnosis:Sex + Age:Sex + Diagnosis:Age:Sex
In other words, all the m
by
gang
-
AFNI Message Board
> Does the false discovery rate (FDR) correction in AFNI refer to the Benjamini-Hochberg procedure?
Yes, the main principle is based on the paper:
Benjamini Y, Hochberg Y (1995). Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society, Series B. 57(1): 289–300.
by
gang
-
AFNI Message Board
> dose it mean that all these pb04.Pilot1.r01.scale+orig.HEAD should align to the same 3DT1 space,then use the 3dDeconvolve regress the 6*5=30 scaled files.
Ideally all the data should be aligned to the same template in the standard space before you feed them to the subject-level regression model.
by
gang
-
AFNI Message Board
> way to tell how much might be needed?
Trial-level modeling is quite demanding on memory size. You should be able to estimate the full memory size based on each subject's file size.
> Any suggestions?
Are the input data stored in the output from subject-level regression analysis? Estimate the total input size, and see if your computer has enough memory to handle that. If so
by
gang
-
AFNI Message Board
> Is it possible that 3dttest++ takes into account all the subjects in a covar file, even those without a matching 3d input file?
Bob might have the specific answer about this, but your test seems to indicate that 3dttest++ cannot properly parse the covariate file when more data are included than needed.
by
gang
-
AFNI Message Board
I'll try to summarize your data structure and see if I understand it accurately. You have
2 groups of subjects
each subject has 2 effects/conditions
> we wanted to study the correlation between the Reaction Time and ReHo task-ReHo rest in SCH and TYP (within-group analysis)
Is Reaction Time associated with the task condition only or both task and rest? You have the contrast betw
by
gang
-
AFNI Message Board
If all the data across sessions are spatially aligned, you can put all the data for each subject into one regression model through afni_proc.py or 3dDeconvolve and obtain one effect estimate per taste at the subject level.
by
gang
-
AFNI Message Board
Again, could you clarify your data structure: Does each subject have one or two 3D input files? Also, what are the research questions you would like to address?
by
gang
-
AFNI Message Board
Marie,
First, follow examples 5, 9 or 10 in afni_proc.py to perform seed-based correlation analysis for each subject. Then, use 3dttest++ or 3dMVM for ANCOVA.
by
gang
-
AFNI Message Board
So, you have 36 trials for each taste? It is very difficult for me to make suggestions when the information your provided is not clear to me. What is your ultimate goal: compare the 5 tastes at the group level, ICC estimation, or something else?
Also, this thread is for the program test-rest reliability. So, could you start a new thread and provide more complete information?
by
gang
-
AFNI Message Board
> ATTEMPT1: I created a unique covar file (CFILE_ALL), which included all 139 subjects and their RT (a file of 2 columns and 140 rows),
> because I thought that subjects not included in the 3dttest++ script would have been ignored.
There are two groups, and each subject has two conditions, right? Does each subject have two 3D input files? If so, I don't see how 3dttest++ script cou
by
gang
-
AFNI Message Board
007-
> it sounds like following this advice seems to be good for 3dlmer, too. Is that correct?
In my opinion, the reporting suggestions discussed in that preprint do apply to most scenarios in neuroimaging, regardless of the modeling framework.
The massively univariate modeling approach (i.e., voxel-wise analysis) assumes that the signal follows a uniform distribution across the brain
by
gang
-
AFNI Message Board
Suppose that the effects of the three conditions are A, B, and C. If B < 0, you would have A > B + C and A > B while A < C. For example, (A, B, C) = (2, -1.5, 3) or (-1, -2, 0).
by
gang
-
AFNI Message Board
Jimmy,
The post hoc inferences #10-12 do not make sense when you provide both quantitative variables in the specification. Could you clarify what you mean by those 3 lines?
by
gang
-
AFNI Message Board
> Is there any reason bringing data processed in other software packages over the 3dLMEr might be problematic? For example,
> does 3dLMEr, FWHMx or 3dClustSim make any assumptions about how the data was processed that could be violated by
> data created in other software?
No, I don't see any particular issues as long as the data are reasonably processed and subject-level analy
by
gang
-
AFNI Message Board
Jimmy, try changing
-qvars 'NMfull*TotalUse'
to
-qvars 'NMfull,TotalUse'
Also, are you sure you want to capture the interaction between NMfull and TotalUse through the specification -bsvars 'NMfull*TotalUse' instead of -bsvars 'NMfull+TotalUse'? Lastly, be careful about the centering issue:
by
gang
-
AFNI Message Board
Lisa, the cluster-based approach as a band-aid for mass univariate modeling would be equally applicable among all the post hoc inferences since they share the same spatial smoothness parameters.
by
gang
-
AFNI Message Board
I was just responding to your previous message while your update came in! Yes, you already found out the problem: when MEdu = 2, there are no Female subjects in the LR group:
Sex
Group Female Male
HR 7 12
LR 0 3
> Is there a way to still include Sex as a factor without the program erroring out?
Unfortunately you cannot have both Sex and MEdu in th
by
gang
-
AFNI Message Board
With such as small subject sample size as 4 subjects, it could be challenging to perform group analysis as well as reliability estimation.
For conventional ICC, you can obtain the effect estimates for each task, and feed them into 3dICC.
by
gang
-
AFNI Message Board
Unfortunately it is difficult for me to understand your goal and your data structure.
1) Are you trying to calculate the ICC at the whole brain or region level?
2) ICC for each task separately or for some contrast among the 5 tasks?
3) "each session has 6*6*5=180 trails" - what are the two 6s?
by
gang
-
AFNI Message Board
> Reason: Incompatible library version: R_io.so requires version 3.6.0 or later, but libR.dylib provides version 3.4.0
Update your R version from 3.4.0 to 3.6.x (not 4.x), and then try it again.
by
gang
-
AFNI Message Board
Stef,
> Specifying an initial p-value of 0.001 cluster-extend thresholded (k=32) to achieve alpha = 0.05,
> I find effects of group and the covariate, but no interaction.
The conventional cluster-based adjustment for multiple testing can be excessively conservative ( ). So, try setting voxel-wise p-value to 0.05 and see if you could identify some interesting results.
> The pro
by
gang
-
AFNI Message Board
> For our purposes, would it be best to model the categorical interaction using 3dMVM?
Yes, 3dMVM would be the easiest program for this case.
by
gang
-
AFNI Message Board
Next time when you design an FMRI/behavioral experiment, perhaps give a little bit more attention to trial sample size?
The following is the abstract of our recently published paper regarding the importance of trial sample size:
Here we investigate the crucial role of trials in task-based neuroimaging from the perspectives of statistical efficiency and condition-level generalizability. Big
by
gang
-
AFNI Message Board
Janelle, your model specification looks fine if you don't believe there is interaction between Group and FatherAge.
by
gang
-
AFNI Message Board
Janelle,
> is the reason behind why dummy coding is not necessary for 3dMVM because the program can accept explanatory variables
> that are categorical (i.e., factors) in addition to explanatory variables that are quantitative?
The user does not need to dummy code categorical variables because 3dMVM (and other programs such as 3dLME and 3dLMEr) dummy code them internally.
> Wou
by
gang
-
AFNI Message Board
Austin,
You can use 3dttest++ to perform a simple regression analysis. If you want the correlation coefficient, use the formula in following link to covert the t-statistic:
You can use the AFNI GUI 'afni' to visualize the results.
by
gang
-
AFNI Message Board