Release info GraphVar 2.03a – Experimental MATLAB R2026 / Apple Silicon Compatibility Patch:

* added experimental compatibility with MATLAB R2026 and Apple Silicon (ARM64) Macs
* removed deprecated JavaFrame dependencies used for GUI window icons, as JavaFrame is no longer supported in recent MATLAB versions
* added MATLAB-based fallback implementations for tdot, tpinv, tlstsq, and trandlstsq, allowing these functions to run on Apple Silicon without the corresponding precompiled Intel MEX binaries
* the MATLAB fallback implementations are experimental and have not yet been fully validated against the original MEX implementations
* trandlstsq uses MATLAB’s random number generation instead of the C++ random number generator used by the original MEX implementation; therefore, identical randomization sequences should not be expected
* other legacy MEX functions have not yet been ported and some GraphVar functionality may therefore still be incompatible with MATLAB R2026 on Apple Silicon
* this patch should be considered experimental and is not yet intended as a fully validated replacement for the original GraphVar 2.03a release

Release info GraphVar 2.03a:

- removed a bug for the selection of the within subj ID buttom (callback went missing in the previous release - sorry!)

Release info GraphVar 2.03:
 
- included network construction methods as provided in BrainNetClass (i.e., high-order FC,  FC based on sparse representations, FC based on group sparse representations)
- when using these embedded methods please cite the corresponding paper from Zhou et al., 2020

Release info GraphVar 2.02b:
 
- removed a bug for the case of calculation on raw matrices with parametric p-values and FDR correction that entered the last release. For some scenarios this resulted in too liberal control of false positives.
Other settings were not affected
- removed a bug for local path lenght (global was claulated instead)


Release info GraphVar 2.02:
 
- added community functionalities as in Fornito et al., 2012: https://www.pnas.org/content/109/31/12788
- added dynamic network measures as in Seizemore et al., 2017: https://www.sciencedirect.com/science/article/pii/S1053811917305645
- fixed a bug when displaying R2 values (issues with the X axis)
- added 'Result', '-v7.3' to execute stats (i.e., now any size of result can be saved)
- fixed bug when running permutation analyses with 3 between factors (worked previously and got lost in some updates)
- fixed bug when running GraphVar on Retina displays



Release info GraphVar 2.01c:

- fixed an issue for LinSVM (classification, regression, probabilisitc): tuned hyperparameters derived from nested-cross validation
were not applied to the models (i.e., prediction was similar to no hyperparameter optimization). ElasticNet was unaffected.

Release info GraphVar 2.01b:

- fixed an issue with manual hyperparameter entry for elastic net (... wrong GUI handle, did not affect results)
- fixed an issue when diplaying p-values of negative weights in the machine learning results viewer
(this was just a visaul issue that did not affect classification/regression results).

Release info GraphVar 2.01:

- added a new compiled MEX file for use on MAC with newer Matlab versions
- Matlab 2018 colorbar compatibility issue resolved
- fixed bug "missing brainsheet" in SampleWorkspace 

Release info GraphVar 2.0:


Background: We previously presented GraphVar as a user-friendly
MATLAB toolbox for comprehensive graph analyses of functional
brain connectivity. Here we introduce a comprehensive extension of
the toolbox allowing users to seamlessly explore easily customizable
decoding models across functional connectivity measures as
well as additional features.

New Method: GraphVar 2.0 provides machine learning (ML)
model construction, validation and exploration. Machine learning
can be performed across any combination of network measures
and additional variables, allowing for a flexibility in neuroimaging
applications.

Results: In addition to previously integrated functionalities, such
as network construction and graph-theoretical analyses of brain
connectivity with a high-speed general linear model (GLM), users
can now perform customizable ML across connectivity matrices,
network metrics and additionally imported variables. The new
extension also provides parametric and nonparametric testing of
classifier and regressor performance, data export, figure generation
and high quality export.

Comparison with existing methods: Compared to other existing
toolboxes, GraphVar 2.0 offers (1) comprehensive customization,
(2) an all-in-one user friendly interface, (3) customizable model
design and manual hyperparameter entry, (4) interactive results
exploration and data export, (5) automated cueing for modelling
multiple outcome variables within the same session, (6) an easy to
follow introductory review.

Conclusions: GraphVar 2.0 allows comprehensive, user-friendly
exploration of encoding (GLM) and decoding (ML) modelling
approaches on functional connectivity measures making big data
neuroscience readily accessible to a broader audience of neuroimaging
investigators.




---> there is a preprint version of a corresponding new GraphVar ML articel on arxiv.org  <---