Multicenter comparison of sensory evoked fMRI in the rat.

This is the preprocessed data for the multirat sensory evoked project (see preregistration: 10.17605/OSF.IO/8VY9R). We aggregated 220 scans from 12 laboratories from around the world. The raw data is available in the BIDS format on openneuro.org (10.18112/openneuro.ds005534.v1.0.0) . The data was preprocessed using RABIES and modelled using Nilearn. The code to preproduce the analysis is freely available on GitHub as Jupyter notebooks (https://github.com/grandjeanlab/multirat_se). This collection contains the first-level analysis with a General Linear Model (GLM) for the functional MRI scans using 4 haemodynamic response functions (hrf) and 1 box model. The first-level data is organized a: ./first_level/[hrf]/beta_estimates/[subject_ID]: the beta parameter estimate of the GLM, mapped onto the brain. ./first_level/[hrf]/z_score/[subject_ID]: the z-score of the GLM, mapped onto the brain. ./first_level/[hrf]/image/[subject_ID]: the z-score shown as a colour-coded overlay over the mouse brain. This collection also contains the second-level analysis General Linear Model for the group analysis per model. The second-level data is organized similarly, but at the group (dataset) level