Mengbo-Li/brainClass: Transcriptional Data Guided fMRI Network Classification

The brainclass R package is a toolkit that performs transcriptional -data-guided fMRI network classification to link connectome with the brain transcriptome. Via the strategic construction of feature groups which incorporates biologically meaningful gene set information such as biological pathways or gene ontology terms, the brainclass framework links predictive functional connectivity features with the potentially involved biological pathways, to bridge the gap between the functional biomarkers of neurological disorders and their underpinning molecular mechanisms. In addition, we also provide a post-hoc interpretation framework to enable biologically meaningful interpretation for the functional edge selection results by other existing fMRI network classification algorithms. The proposed methods can be easily applied to task-based fMRI data, and the selection of brain parcellation for brain network construction is entirely flexible.

install.packages("remotes")
remotes::install_github("Mengbo-Li/brainClass")