AFNI Notes: 6/23/2016

 Description Comment
 3dttest++  a relatively flexible voxel based t-test
 really good option for obvious contrasts
           3dttest  as above without new features
 largely decommissioned
           3dMEMA  like 3dttest++ but uses t-stat for scaling
 needs REML, conceptually cool, not sure of benefit
 non parametric 2 group ttest non-paired
 maybe if the data is just that noisy 
           3dKruskalWallis    as above for >2 groups
or multiple noisy datasets
 paired 2 group non parametric
 as above for >2 groups
 3dLME  a nice general LM frame work
 good option for interactions etc.
            3dMVM  like 3dLME with a few less bells
 if your comfortable with LME just stay with that
            3dANOVAx  Does x way ANOVAs
 as it needs equal groups and other reasons don't use
            3dRegAna  Does multiple regression
 as it does calculations in short and other reasons don't use
3dttest++             Student T test with 1 or 2 sets of data (labeled 'A' and 'B').
-prefix p               p = output
-mask mmm        mmm = specified mask.
-setA/-setB         Use this to specify group membership
1 group just give setA to contrast with base 0 [or setB versus -singletonA 10/perfect10+tlrc'[1]']
2 group will give setA-setB, setA(v. null), and setB(v. null) [-BminusA flips contrast]
Short form: -labelA Group1 -setA mydatasub1+tlrc'[1]' mydatasub2+tlrc'[1]'  … etc.
Long form: -setA Group1 sub1 mydatasub1+tlrc'[1]' sub2 mydatasub2+tlrc'[1]'  … etc.

-covariates COVAR_FILE a text file with a table for numeric covariate(s)[columns] for each subj[rows]
Rows must match dataset label example:
                  subject IQ   age  GMfrac
                  Elvis   143   42  Elvis_GM+tlrc[8]
                  Fred     85   59  Fred_GM+tlrc[8]
Ethel   109   49  Ethel_GM+tlrc[8]
Lucy    133   32  Lucy_GM+tlrc[8]
Ricky   121   37  Ricky_GM+tlrc[8]
can only be used with the short form '-setX' option if each input dataset has only 1 sub-brick
-center NONE/DIFF/SAME             do you remove mean of covariates? Group or set mean? [default DIFF]
-paired                 is the data paired. If paired setA will be ignored
-pooled/unpooled          for un/pooled variance (blocks covariates)
-toz                       Convert output t-statistics to z-scores
-rankize                Convert the data (and covariates, if any)
-brickwise            run ttest on each subbrick for each brick provided

-resid q                 q = Output residuals to look at with 3dFWHMx
-randomsign       makes better residuals by flipping directionality
-clustsim              will run  10000 iteration clustsim and apply results to file
-Clustsim             keep .1D files
-CLUSTSIM          keep all temp files
-prefix_clustsim cc           cc = prefix of clustsimfiles

3dLME               Multivariate Linear Model
-jobs NJOBS     processors
-prefix               Output file
-mask MASK     Process voxels inside this mask only.
-model FORMULA
        Test of model like '~A*B+C+A:B” + = factors; : = interaction * = both (A*B = A+B+A:B)
-SS_type NUMBER        Specify the type for sums of squares: 1=sequential 3=marginal
-qVars variable_list       comma separated list of quantitative variables
-qVarCenters VALUES   comma separated list of means of quantitative variables [default = group mean]
-vVars variable               single voxel-wise covariate, i.e. subXTime1+tlrc column in dataTable
-vVarCenters VALUE     centering values for voxel-wise covariates [default=mean of all subjects]
-ranEff FORMULA          Specify the random effects. For just Subjects = "~1" but could be"~1+RT"
-num_glf/num_glt NG  number of general linear F-tests/t-tests
-glfCode/gltCode k CODING       Specify the k-th general linear F-test/t-test
Coding example:  'Condition : 1*A -1*B & 1*A -1*C Emotion : 1:pos'

  • Weights don’t need to sum to 0
  • If covariate is not given the mean is used
  • If a categorical is not given the mean of all levels I used

   -glfLabel/gltLabel k label             Specify the label for the k-th GLF/GLT
      List the data structure with a header as the first line.

  • Last item command
  • Header on first line must start with Subj and end with InputFile
  • Can be a file '-dataTable @table.txt'
Other interesting:
-corStr FORMULA          Correlation structure of the residuals (like AR or ARMA) if timeseries entered.
-cutoff threshold: Specify the cutoff value to obtain voxel-wise accuracy [default is 0]
-ICC    intra-class correlation [for variables in –ranEff]
-LOGIT              Do logistic regression [no random effects]