AFNI HISTORY

The levels of importance go from 1 to 5, with meanings:
       1 - users would not care
       2 - of little importance, though some users might care
       3 - fairly important
       4 - a big change or new program
       5 - IMPORTANT: we expect users to know

generated via the command : afni_history -html -reverse -min_level 4 -past_years 2 

  ----  log of AFNI updates (most recent first)  ----

20 May 2009, G Chen, 3dMEMA.R, level 4 (SUPER), type 1 (NEW_PROG)
    Mixed-Effects Meta Analysis (MEMA)

    This is an R program that runs group analysis in a truly random
    mixed-effects sense by taking both beta and t-statistic as input
    instead of beta value only in the conventional method. See more 
    details at http://afni.nimh.nih.gov/sscc/gangc/MEMA.html


24 Oct 2008, RC Reynolds, xmat_tool.py, level 4 (SUPER), type 1 (NEW_PROG)
    program to inspect a .xmat.1D X-matrix, possibly against a time series

    This is a Graphical tool for plotting a design matrix, reviewing
    condition numbers or the correlation matrix, and fitting to a 1D
    time series.


09 Oct 2008, G Chen, 1dGC.R, level 4 (SUPER), type 1 (NEW_PROG)
    Multivariate Granger causality analysis

    This is an R program that runs Granger causality analysis among a few
    pre-select regions. See more details at
    http://afni.nimh.nih.gov/sscc/gangc/1dGC


21 Aug 2008, RW Cox, 3dREMLfit, level 4 (SUPER), type 1 (NEW_PROG)
    Program to mimic 3dDeconvolve, but with serial correlations

    Uses ARMA(1,1) model of noise, separately for each voxel.


07 Apr 2008, DR Glen, align_epi_anat.py, level 4 (SUPER), type 1 (NEW_PROG)
    Alignment of EPI and Anatomical datasets

    Aligns anat and EPI data. Alignment is in either direction of anat to
    epi or epi to anat. Transformations are combined where possible as
    from time series registration and talairach transformations. Multiple
    child epi datasets may be aligned at the same time.


29 Feb 2008, G Chen, 3dICA.R, level 4 (SUPER), type 1 (NEW_PROG)
    Independent component analysis

    This is an R program that runs independent component analysis. It
    takes one dataset that presumably has already been properly
    pre-processed.already been properly pre-processed. See more details at
    http://afni.nimh.nih.gov/sscc/gangc/ica.html


27 Feb 2008, RC Reynolds, afni_history, level 4 (SUPER), type 1 (NEW_PROG)
    program to display the history of AFNI updates

    This will be used to create a web page of AFNI updates.
    Please see 'afni_history -help' for more details.


21 Feb 2008, RC Reynolds, GIFTI, level 4 (SUPER), type 0 (GENERAL)
    AFNI programs can now read and write GIFTI datasets

    GIFTI datasets are for data in the surface domain, with file suffix .gii.
    Support must be requested at compile time, and it requires libexpat.
    Please see http://www.nitrc.org/projects/gifti for many details.


16 Feb 2008, RW Cox, 3dTfitter, level 4 (SUPER), type 1 (NEW_PROG)
    new program = linear fits to voxel time series

    Uses L1 or L2 regression, with optional constraints to fit each voxel
    time series as a sum of basis time series, which can be 1D files or
    3D+time datasets.  Basis time series that are 1D time series are
    the same for all input voxels.  Basis time series that are 3D+time
    datasets are different for each voxel.
    Differences from 3dDeconvolve:
    * Basis time series can vary across voxels.
    * Fit coefficients can be found with L1 or L2 error functions, and
      can be constrained to be positive or negative.
    * 3dTfitter does not compute goodness-of-fit statistics.