An R package for performing graph theory analyses of brain MRI data from structural, DTI, and resting-state fMRI connectivity.
brainGraph is an R package that applies graph theory to analyze brain MRI data, modeling the brain as a network of interconnected regions. It computes various graph metrics to study brain connectivity patterns and differences across groups or conditions. The package supports multiple MRI modalities including structural, DTI, and resting-state fMRI data.
Neuroscience researchers, data scientists, and bioinformaticians who analyze brain MRI data and want to apply graph theory methods to understand brain network organization and connectivity.
It provides a comprehensive, open-source solution within R for graph-based brain network analysis, integrating statistical methods like NBS and MTPC, and supporting numerous standard brain atlases out of the box.
Graph theory analysis of brain MRI data
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Calculates over 20 vertex- and graph-level measures including efficiency, centrality, and small-worldness, leveraging igraph for extensive network analysis capabilities as detailed in the Graph measures section.
Includes data for 14 standard atlases like Desikan-Killiany, AAL, and Harvard-Oxford, enabling immediate analysis without manual atlas integration, with support for custom atlases.
Supports GLM-based group comparisons, Network-Based Statistic (NBS), and Multi-Threshold Permutation Correction (MTPC), providing robust inference tools for connectivity differences, with examples in the User Guide.
Offers a detailed User Guide with code examples and an active Google Group for support, ensuring reproducibility and ease of learning, as highlighted in the Getting Help section.
The visualization GUI requires RGtk2 and cairoDevice, which are notoriously difficult to install on macOS and Windows, limiting accessibility for non-Linux users, as noted in the GUI installation instructions.
Tightly coupled with R and its packages, making it unsuitable for teams using Python or other languages for neuroimaging pipelines without significant workflow changes, despite igraph compatibility.
Longitudinal modeling with linear mixed effects is listed for future versions, so researchers needing time-series analysis must seek alternative tools or wait for updates, as admitted in the Future versions section.