Software
Selection of open-source software contributed by members of the project :
GSDGM and MTESS Toolbox
Group Surrogate Data Generating Model (GSDGM) and Multivariate Time-series Ensemble Similarity Score (MTESS) Toolbox for MATLAB.
The GSDGM and MTESS Toolbox is a powerful tool for surrogate data generation and multivariate time-series similarity analysis.
(Python version is also available on github).
Contributor: Connectome Analysis Unit
3D Brain Atlas Viewer
The Brain/MINDS Marmoset Brain 3D Atlas Viewer is now open source!
This 3D viewer has been developed using Unity engine, and the provided sources contain scripts and 3D region models of the Marmoset brain generated from Brain/MINDS 3D Marmoset Reference Brain Atlas.
Contributor: Connectome Analysis Unit
AbART - ANTs based Atlas Registration Tool
AbART is a web based interactive tool to register MRI volumes to Brain/MINDS Marmoset Atlas.
Users can load and visualize their MRI volumes (NIfTI-1 format), interactively reorient them (to allow correct registration), then submit for registration & transformation, and finally retrieve their image volume transformed in the Atlas space.
Contributor: Connectome Analysis Unit
Predictive principal component analysis (PredPCA)
This is the MATLAB code for PredPCA, an analytically solvable unsupervised learning scheme that extracts the most informative components for predicting future inputs (see Nature Machine Intelligence).
Contributors: Takuya Isomura, Brain Intelligence Theory Unit , Taro Toyoizumi, Neural Computation and Adaptation
Simulations of spiking neural networks with reward-dependent local synaptic plasticity rules and traveling waves
This is the supporting code for "Learning poly-synaptic paths with traveling waves" paper (see PLOS Computational biology) that was used to perform simulations (leveraging Brian2 library) and their post-analysis.
Contributors: Yoshiki Ito, Taro Toyoizumi, Neural Computation and Adaptation
Supporting code for "A Bayesian psychophysics model of sense of agency"
This is the supporting code for "A Bayesian psychophysics model of sense of agency" paper (see Nature Communications) that was used to generate the simulation data and plot the computation results for analyses.
Contributor: Taro Toyoizumi, Neural Computation and Adaptation
Oscillator decomposition of time series data (OSC-DECOMP)
This is the MATLAB code for OSC-DECOMP, a statistical method for extracting oscillators from univariate or multivariate time series data.
As an example related to brain study, this code has been used to investigate oscillatory activity in functional near-infrared spectroscopy (fNIRS) data.
(The fNIRS can detect hemodynamic responses in the brain, where collected data are bivariate time series of oxygenated and deoxygenated hemoglobin, see PLOS Computational Biology, 2022 for details about this specific application).
Contributor: Takeru Matsuda, Statistical Mathematics Unit
Vector Auto-Regressive Deep Neural Network (VARDNN) toolbox
VARDNN is a powerful tool of data-driven analysis technique to estimate directed FC (Functional Connectivity). Based on VARDNN framework, two types of directed FC are defined, such as VARDNN-DI and VARDNN-GC to measure causal relation among multiple time-series data.
This toolbox includes several functional connectome measures, such as VARDNN-DI, VARDNN-GC, VARLSTM-GC, multivariate Granger Causality (GC), pairwise GC, multivariate Principal Component (PC)-GC, multivariate Partial Least Squares (PLS)-GC, multivariate Elastic Net (EN)-GC, linear Transfer Entropy, Functional Connectivity (Correlation), Partial Correlation, PC-PC, PLS-PC, EN-PC and Wavelet Coherence to estimate conectivity from multiple node signals.
Update: VARDNN has been ported to Python and is available on github as well.
Contributor: Connectome Analysis Unit
ZAViewer - Zooming Atlas Viewer
ZAViewer is a web 2D image viewer that was primarily designed to explore the Brain/MINDS Marmoset Reference Atlas.
It can display up to 3 sets of multimodal, regularly interspaced, large image slices along the 3 standard orthogonal axes (Axial, Coronal, Sagittal). Each slice view may contains several raster images layers, and Atlas regions (represented by aligned vector images, SVG) displayed as an overlay over the raster images.
Contributor: Connectome Analysis Unit