﻿About SDMPsi v6.23 software
========================

Seed-based d Mapping, Permutation of Subject Images software 
is free to use and distribute as long as this file's credits are maintained. 
Please get in touch with Joaquim Radua or Anton Albajes-Eizagirre at 
http://www.sdmproject.com/ if you have any doubts.

This software was created and has been developed by Joaquim Radua and
Anton Albajes-Eizagirre with the kind support of David Mataix-Cols and
other collaborators. Please cite the papers in which the SDMPsi method is
described:

Introduction to voxel-based meta-analyses:
1. Radua J and Mataix-Cols D. Meta-analytic methods for neuroimaging 
data explained. Biol Mood Anxiety Disord 2012; 2:6. 

2. Müller VI et al. Ten simple rules for neuroimaging meta-analysis. 
Neurosci Biobehav Rev 2018; 84:151-161. 
SDM methods:

First method (SDM): 
3. Radua J and Mataix-Cols D. Voxel-wise meta-analysis of grey matter 
changes in obsessive-compulsive disorder. Br J Psychiatry 2009; 195:393-402. 

4. Meta-comparisons: Radua J, van den Heuvel OA, Surguladze S, and Mataix-Cols D. 
Meta-analytical comparison of voxel-based morphometry studies in obsessive 
compulsive disorder vs other anxiety disorders. Arch Gen Psychiatry 2010; 
67:701-711. 

5. Effect sizes (ES-SDM): Radua J, Mataix-Cols D, Phillips ML, El-Hage W, 
Kronhaus DM, Cardoner N and Surguladze S. A new meta-analytic method for 
neuroimaging studies that combines reported peak coordinates and statistical 
parametric maps. Eur Psychiatry 2012; 27:605–611. 

6. Anisotropic kernels (AES-SDM): Radua J, Rubia K, Canales-Rodriguez EJ, 
Pomarol-Clotet E, Fusar-Poli P and Mataix-Cols D. Anisotropic kernels for 
coordinate-based meta-analyses of neuroimaging studies. Front Psychiatry 2014; 
5:13. 

7. [issues of previous methods] Test for spatial convergence: Albajes-Eizagirre A 
and Radua J. What do results from coordinate-based meta-analyses tell us? 
Neuroimage 2018; 176:550-553. 

8. [main paper of the current method] Permutation of subject images (SDM-PSI): 
Albajes-Eizagirre A, Solanes A, Vieta E, and Radua J. Voxel-based meta-analysis 
via permutation of subject images (PSI): Theory and implementation for SDM. 
Neuroimage 2018; 186:174-184. 

SDM-specific white matter and TBSS templates:
9. White matter: Radua J, Via E, Catani M, and Mataix-Cols D. Voxel-based 
meta-analysis of regional white-matter volume differences in autism spectrum 
disorder versus healthy controls. Psychol Med 2011; 41:1539-1550.  link
10. Fractional anistropy (TBSS): Peters BD, Szeszko PR, Radua J, Ikuta T, Gruner P, 
DeRosse P, Zhang JP, Giorgio A, Qiu D, Tapert SF, Brauer J, Asato MR, Khong PL, 
James AC, Gallego JA, and Malhotra AK. White matter development in adolescence: 
Diffusion tensor imaging and meta-analytic results. Schizophrenia Bull 2012; 
38:1308-1317. 

Improved atlas:
11. Radua J, Grau M, van den Heuvel OA, de Schotten MT, Stein DJ, 
Canales-Rodriguez EJ, Catani M, and Mataix-Cols D. Multimodal voxel-based 
meta-analysis of white matter abnormalities in obsessive-compulsive disorder. 
Neuropsychopharmacology 2014; 39:1547-57. 

12. Thiebaut de Schotten M, Dell'Acqua F, Forkel SJ, Simmons A, Vergani F, Murphy DG 
and Catani M. A lateralized brain network for visuospatial attention. Nat Neurosci 2011; 
14:1245-1246. 

13. Thiebaut de Schotten M, Ffytche DH, Bizzi A, Dell'Acqua F, Allin M, Walshe M, 
Murray R, Williams SC, Murphy DG and Catani M. Atlasing location, asymmetry and 
inter-subject variability of white matter tracts in the human brain with MR diffusion 
tractography. Neuroimage 2011; 54:49-59. 

14.  Cortical thickness (FreeSurfer): Li Q, Zhao Y, Chen Z, Long J, Dai J, Huang X,
Lui S, Radua J, Vieta E, Kemp GJ, Sweeney JA, Li F and Gong Q. Meta-analysis of
cortical thickness abnormalities in medication-free patients with major depressive
disorder. Neuropsychopharmacology 2019; in Press.

SDM software log
================
v6.23 - New multimodal computation bugs fixed or Sdm-Psi.
(02/24)	Added missing libpng on some Linux distributions.
        Fixed selection of the number of threads to be used in the  GUI.
	Fixed bug on cluster extension on GUI.
	Duplicated studies in SDM table check added.
	Bug fixes on FWE for interceptless analysis.
	Bug fixes on FWE for analysis without enough studies for the selected permutations.

v6.21 - New FreeSurfer mask.
(11/19) New multimodal implemented.
	New feature on FWE: if previous permutations are computed, reuse them and
	compute only additional required (i.e., if perm was stopped before completing, resume).
	Bug fixes on SDM table and bias test.
	Bug fixes for memory leaks in FWE.
	Bug fixes 'n' value is used as the peak value.
	Bug fixes for GUI's batch processing for Mean and FWE.

v6.12 - Bug fixes for analyses with NAs present in vars.
(07/19) Bug fixes for SDM table wider than higher (more vars than studies).
	Bug fixes for batch runs.
	Bug fixes for imgcalc in Windows platforms.
	Other minor bugs were fixed.

v6.11 - NEW METHOD. See reference #8 Albajes-Eizagirre2018 at the top of this document.
(04/19) * (nearly) Unbiased estimation of effect sizes based on MetaNSUE algorithms
	* Familywise correction for multiple comparisons using common permutation tests, 
	i.e., permuting subject images (PSI). Note that previous methods used tests
	for spatial convergence instead, which rely on spatial assumptions that may not 
	be met and have lower power in the presence of multiple effects.

	* Freedman-Lane-based permutation, for its optimal statistical properties  
	* Threshold-free cluster enhancement (TFCE) statistics, which were neither too 
	conservative nor too liberal in the simulation work (whereas voxel-based statistics 
	were too strict and cluster-based statistics too liberal).

	*…  and some other improvements (Disclaimer: do not hold any key pressed while 
	closing the ‘About’ dialog of the GUI ;-) 

v5.15 - Moderate changes version. Bug fix for Multimodal analysis crashing
(04/18) in Windows.
	Changed labels to make more explicit differences between Simplified
	Log and Execution Log. 

v5.14 - Bug fixes for missing Peak Maps creation at Threshold on core binaries
(10/16) Bug fixes for multimodal.
	Some bug fixes on OSX (open .nii.gz files dialogs)

v5.13 - Bug fixes for Combine Images on core binaries
(09/16)

v5.12 - Bug fixes for Create Mask on core binaries. 
(08/16)

v5.11 - Major changes version. Main improvements: New improved graphical
(08/16) user interface for Windows, Linux, and Mac OSX, possibility to
        combine repeated measures (e.g., from several contrasts of the
        same sample), easy automatic creation of funnel plots and Egger
        tests. Other improvements: recreated t-maps are also saved (along with
        the effect size maps), threshold procedures create a mask for each peak
        (rather than for each blob), added "other_brett" for FreeSurfer. Solved
        a small bug in the automatic creation of masks when thresholding SDM
        results (masks were sometimes empty).

v4.31 - Moderate changes version. Main changes: specific MNI transforma-
(05/15) tions have been introduced for FSL and software other than SPM
        (to avoid confusion, *.mni.txt must now be one of:
        *.spm_mni.txt, *.fsl_mni.txt or *.other_mni.txt; *.mni2tal.txt
        must be now one of: *.spm_brett.txt or *.fsl_brett.txt;
        *.tal.txt must be *.other_tal.txt; and any of them may be
        *.no_peaks.txt if it includes no peaks), threshold procedures
        (in both imgcalc and SDM) create a mask for each blob (which may
        be used to extract mean blob values), *_QH_z and *_Q2_z maps are
        now called *_z_QH and *_z_Q2, and multimodal analyses can now be
        run from the command line and optionally conducted assuming no
        error in p-values, without reducing false positive rates, and
        specifying the tolerable p-values from the unimodal meta-analy-
        ses.

v4.22 - Minor changes version. Solved a bug related to thresholding in
(05/15) multimodal meta-analyses.

v4.21 - Major changes version. Main improvements: the software is now
(01/15) also compiled for Apple OSX (though currently without GUI), in-
        clusion of the white matter atlas by Catani, Thiebaut de Schot-
        ten et al. (which includes structures not present in other atlas-
        es such as the U and aslant tracts), and thresholding has been
        recoded to increase speed and be called from imgcalc (which now
        includes five new thresholding/blobs commands: -atlas for at-
        las information, -blobs for simple blob information, -ethr for
        cluster extent thresholding, -sdm for atlas and SDM information,
        and -pthr for cluster peak thresholding). Other improvements:
        detailed console output for imgcalc, sdm_main.nii.gz file has
        been replaced by individual *.nii.gz files, the multimodal algo-
        rithm has been rewritten to speed up the analyses, the SDM table
        may have empty cells (which are automatically filled with NA)
        and a column called jackknife without throwing errors, and the
        GUI checks that filenames do not have forbidden characters and
        sorts dialog lists. Solved two minor bugs: in some systems, the
        program asked for updates even if the last version was already
        installed, and I2 was not correctly printed. Other changes:
        thresholding does not create logp maps anymore (these can be
        easily obtained from p maps using imgcalc), "invp" has been re-
        placed by "p" in imgcalc.

v4.13 - Minor changes version. Solved two small bugs: the program froze
(03/14) in Windows when checking updates depending on the internet con-
        nection, and Brodmann areas were occasionally wrongly displayed.

v4.12 - Minor changes version. Solved bugs that made the thresholding
(02/14) crash if no threshold was specified or if peaks were close to
        the end of the volume.

v4.11 - Major changes version. Main additions: Pre-processing of peaks
(02/14) is conducted taking into account the anisotropy in the spatial
        covariance, new masks, and correlation templates have been added
        (e.g. cerebrospinal fluid), many parts of the software have been
        rewritten to increase compatibility (e.g., use of MNI instead of
        Talairach space, or *.nii.gz and *.txt instead of *.sdm files,
        so that users may use their templates, masks, and etcetera),
        one-sample studies (e.g., correlation with a measure) and two-
        sample studies (e.g., high vs. low measure) may be combined in a
        single meta-analysis. Other improvements: heterogeneity I2
        statistic is also reported, command-line SDM / imgcalc work in the
        directory from where they are called (i.e., two meta-analyses can
        be safely conducted at the same time), automatic updates, Win-
        dows version is also compiled in 64 bits, many parts of the code
        have been rewritten to optimize performance and avoid memory
        overloads, and bugs in the specification of the FSLView executable
        and multimodal folders have been addressed. Other changes:
        Talairach labels have been replaced by AAL (with names standardized
        according to the Online Brain Atlas Reconciliation Tool),
        JHU (provisional) and Brodmann areas, masks can be created from
        the command line with "mask", and it's not necessary to specify a
        name for the masks), pp parameters have changed, isotropic FWHM
        is set in the pp dialog (rather than in settings), the old ori-
        ginal method and the interactive console interface have been de-
        precated, and the commands quantile and citation have been de-
        precated.

v3.11 - Major changes version. Main addition: incorporation of the mul-
(08/12) timodal analysis. Other modifications: multiple regression can
        be directly set from the menus, threshold of analyses with no
        findings also returns images.

v2.31 - Moderate changes version. Additions: incorporation of the TBSS
(04/12) template to conduct meta-analyses of TBSS diffusion studies and
        imgcalc also conducts logarithmic, power, threshold, trigonome-
        tric and other basic operations. Improvement: a bug from the version
        2.14 that caused p- and logp-images to sometimes include less
        voxels than those reported has been solved.

v2.21 - Moderate changes version. Improvements: output reports are dis-
(03/12) played in user-friendly web-like files, simple arithmetic func-
        tions have been added to imgcalc, and reporting of pre-process-
        ing errors is more detailed. Other changes: syntaxis of imgcalc
        has been changed to be similar to that of fslmaths, pre-process-
        ing reports now only include the absolute peaks and SDM pro-
        cessor and interfaces have been further separated to optimize
        memory usage.

v2.14 - Minor changes version. Additions: imgcalc also converts "1-p
(10/11) values" images, sdm_threshold also includes a height threshold
        for z-values as described in the ES-SDM method, and FSLView is
        also supported. Improvements: SDM looks for MRIcron (or FSLView)
        program in standard installation directories, defaults have been
        updated to match the ES-SDM standards, jackknife analysis is
        controlled by the interface, optimizing memory use, images
        produced by sdm_threshold only include the clusters described in
        the text report, sdm_threshold allows spaces in map names and
        sdm_help utility has been replaced by website help.

v2.13 - Minor changes version. Improvements: the "update" library has
(07/11) been replaced to also work in some Linux distributions,
        and a bug in pre-processing has been solved (reading of statis-
        tical images caused the program to stop if there were no negati-
        ve values).

v2.12 - Minor changes version. Improvements: a NIfTI image for each pre-
(07/11) processed study is saved, and a bug in pre-processing has been
        solved (reading of statistical images was set for Analyze rather
        than NIfTI, so that NIfTI files could not be read).

v2.11 - Major changes version. Addition: the new Effect-Size Signed
(05/11) Differential Mapping SDM (ES-SDM) algorithms that combine repor-
        ted peak coordinates and statistical parametric maps. These in-
        clude a pre-analysis of coordinates' effect size, pre-processing
        for NIfTI images, new pre-processing for peaks, and new meta-a-
        nalytic mean / linear model (see Radua et al., Eur Psychiat 2011
        for details). To help the user prepare the data for ES-SDM, con-
        versors to t values for both images and peaks have been included
        in the interface, the former using the new 'imgcalc' utility.
        Other changes: some small interface bugs in the globals proce-
        dure and the SDM table editor have been addressed, NIfTI exports
        have been optimized by including more header parameters, float
        precision, compression (*.nii.gz), and (-log) p-values maps.

v1.31 - Major changes version. Addition: a white matter specific tem-
(11/10) plate has been added to conduct meta-analyses of white matter
        VBM or DTI. Improvements: extract procedure now also outputs to
        the main screen, update checking also works in 64bits Linux.
        Other changes: FDR thresholding has been deprecated.

v1.21 - Major changes version. Improvement: linear model procedure has
(07/10) been expanded to allow comparisons between groups or the use of co-
        variates, batch, and output have been joined in a single-screen
        while the log has been reduced to a secondary screen, the extent
        threshold can be optionally set.

v1.14 - Moderate changes version. Additions: a new procedure (globals)
(04/10) for conducting the meta-analysis of global variables such as the
        global gray matter, and automatic checks for updates. Improve-
        ment: visual style of the graphical interface, linear model varia-
        bles are automatically scaled to have values between 0 and 1,
        citation text.

v1.13 - Moderate changes version. Improvements: batch processing can be
(01/10) called from the menus, and an editor has been included in the
        graphical interface, reporting of local peaks within a cluster,
        SDM core (the main processor) and auxiliary software (table edi-
        tor and help support) have been independized from the interface,
        default parameters in commands, commands can also be called from
        the graphical interface, output appears in a separate window in
        the graphical interface. Other changes: batch commands are an
        option (rather than the main entry) in the console interface,
        several commands have been deprecated (cd, chdir, create.mask,
        dir, mc.weighted.mean, read.table, setwd), two command names
        have been replaced (mc.lm -> lm, mc.mean -> mean), mc.pp command
        action has been included in pp command, additional weighting has
        been deprecated, FDR thresholding (and value thresholding for
        maps other than quartiles) is no longer available in dialogs.

v1.12 - Minor changes version. Additions: extraction of Talairach coor-
(11/09) dinates or labels values, MRIcron brain viewer integration, and
        possibility to run the SDM software as an SPM (Statistical Para-
        metric Mapping) extension. Improvements: cluster breakdown was
        conducted separately for each cluster, Analyze images export was
        replaced by NIfTI (Analyze-compatible) images export, and the
        program correctly reads text files saved in a Mac computer.

v1.11 - First, original version. The program was able to convert coordi-
(07/09) nates to Talairach, preprocess the studies, calculate means with
        jackknife, quartiles, and regressions. It also thresholded
        results by p, FDR, or SDM, and output Analyze images. It came with
        a console version able to batch processing and a GTK graphical
        interface with a table editor for the variables. The program was
        compiled for both Linux and Windows.


About other files
=================

You can find the Qt, 7z, GNU Tar, R, and GNU LIBGSL source code at https://www.qt.io/,
http://www.7-zip.org/, https://cran.r-project.org/sources.html, 
https://www.gnu.org/software/tar/ and https://www.gnu.org/software/gsl/, and several
atlases at MRIcron website (http://people.cas.sc.edu/rorden/mricron/index.html)
or described in:

- Thiebaut de Schotten M, Dell'Acqua F, Forkel SJ, Simmons A, Vergani F,
  Murphy DG, Catani M. A lateralized brain network for visuospatial
  attention. Nat Neurosci 2011;14:1245-1246.

- Thiebaut de Schotten M, Ffytche DH, Bizzi A, Dell'Acqua F, Allin M,
  Walshe M, Murray R, Williams SC, Murphy DG, Catani M. Atlasing location,
  asymmetry and inter-subject variability of white matter tracts
  in the human brain with MR diffusion tractography. Neuroimage 2011;
  54:49-59.
