These are the raw data associated with two publications. Please cite the articles below. 1. F. Deligianni, D.W. Carmichael, Gary H. Zhang, C.A. Clark and J.D. Clayden (2016). [NODDI and tensor-based microstructural indices as predictors of functional connectivity](http://dx.doi.org/10.1371/journal.pone.0153404). *PLoS ONE* 11(4):e0153404. 2. F. Deligianni, M. Centeno, D.W. Carmichael and J.D. Clayden (2014). [Relating resting-state fMRI and EEG whole-brain connectomes across frequency bands](http://dx.doi.org/10.3389/fnins.2014.00258). *Frontiers in Neuroscience* 8:258. ## Description These data were acquired as described in Deligianni et al., 2016. The relevant description is reproduced below. "Imaging data from 17 adult volunteers (11 males, 6 females, mean age: 32.84 +/- 8.13 years) were acquired in a Siemens Avanto 1.5 T clinical scanner using a self-shielded gradient set with maximum gradient amplitude of 40 mT m<sup>-1</sup>. Data were acquired in two sessions: - Simultaneous resting-state EEG-fMRI were acquired with a standard 12 channel coil. The subjects had their eyes open and were asked to remain awake and fixate on a white cross presented on a black background. Subjects were asked to remain still and their head was immobilised using a vacuum cushion during scanning. The fMRI imaging acquisition was based on a T2*-weighted gradient-echo EPI sequence with 300 volumes, TR/TE 2160/30 ms, 30 slices with thickness 3.0 mm (1 mm gap), effective voxel size 3.3 x 3.3 x 4.0 mm, flip angle 75 degrees, field of view (FOV) 210 x 210 x 120 mm. Scalp EEG was recorded during the MRI scan using a 64-channel MR-compatible electrode cap (BrainCap MR, Gilching, Germany) at native frequency of 1000 Hz. The electrodes were arranged according to the modified combinatorial nomenclature, referenced to the FCz electrode. The electrocardiogram (ECG) was recorded, and EEG and MR scanner clocks were synchronised. A T1-weighted structural image was also obtained. - Structural data were acquired with a 32-channel head coil. The NODDI sequence optimisation on the 1.5 T scanner follows the experiment design procedure described by Alexander et al. (*Magnetic Resonance in Medicine* 60:439, 2008), applied to the NODDI model (Zhang et al., *NeuroImage* 61:1000, 2012). The *a priori* model parameter settings are: intracellular volume fraction f = 0.3, 0.5, 0.7, intrinsic diffusivity d = 1.7 x 10<sup>-9</sup> m<sup>2</sup> s<sup>-1</sup>, perpendicular diffusivity set according to the tortuosity constraint (Szafer et al., *Magnetic Resonance in Medicine* 33:697, 1995), Watson concentration parameter kappa = 0.5, 2, 8, 32. The scanner-specific sequence settings are T2 = 80 ms, G<sub>max</sub> = 40 mT m<sup>-1</sup>. Finally, we target a total of 109 measurements to have a total acquisition time of around 16 minutes. The optimisation divides the measurements into three HARDI shells with b = 2400 s mm<sup>-2</sup> (60 noncollinear gradient directions and six b = 0 images), b = 800 s mm<sup>-2</sup> (30 noncollinear gradient directions and three b = 0 images) and b = 300 s mm<sup>-2</sup> (9 noncollinear gradient directions and one b = 0 image). These were acquired with a voxel matrix of 96 x 96, 60 contiguous axial slices, each 2.5 mm thick, FOV 240 x 240 x 150 mm, voxel size of 2.5 x 2.5 x 2.5 mm and TR/TE 8300/98 ms. We have kept the same TE across shells to avoid differences in T2 effects that would need to be accounted for in the diffusion model, and thus we avoided adding complexity. High resolution T1-weighted whole-brain structural images were also obtained in both sessions, with a voxel size of 1.0 x 1.0 x 1.0 mm, TR/TE 11/4.94 ms, flip angle 15 degrees, FOV 256 x 256 x 256 mm, voxel matrix 176 x 216 and 256 contiguous slices. The mean interscan interval was 35 +/- 41.6 days." ## Contents **EEG:** The EEG1 and EEG2 folders contain the EEG data acquired during rest. (They are split up just because of their large size.) This includes the "raw" data acquisition obtained directly with the BrainCap MR, located in a folder called raw for each subject. For each subject there is also a folder called export, which contains the EEG data after preprocessing using Brain Vision Analyzer 2. Preprocessing is described in details in the publications above. **fMRI:** The fMRI folder contains the fMRI data during rest (3_nw_mepi_rest_with_cross.nii.gz) and the high-resolution T1-weighted images obtained during the same session, which have had the face region masked out for anonymity (2_fl3D_t1_sag_fun_defaced.nii.gz). **NODDI:** The NODDI folder contains the NODDI acquisition data: - *_NODDI_66DIR_B2400.nii.gz: HARDI shell with b = 2400 s mm<sup>-2</sup> (60 noncollinear gradient directions and six b = 0 images) - *_NODDI_66DIR_B2400.bvals: Corresponding b-values for each gradient direction - *_NODDI_66DIR_B2400.bvecs: Corresponding vectors for each gradient direction - *_NODDI_33DIR_B800.nii.gz: HARDI shell with b = 800 s mm<sup>-2</sup> (30 noncollinear gradient directions and three b = 0 images) - *_NODDI_33DIR_B800.bvals: Corresponding b-values for each gradient direction - *_NODDI_33DIR_B800.bvecs: Corresponding vectors for each gradient direction - *_NODDI_10DIR_B300.nii.gz: HARDI shell with b = 300 s mm<sup>-2</sup> (9 noncollinear gradient directions and one b = 0 image) - *_NODDI_10DIR_B300.bvals: Corresponding b-values for each gradient direction - *_NODDI_10DIR_B300.bvecs: Corresponding vectors for each gradient direction This folder also contains a high-resolution T1-weighted image obtained during the same session (2_fl3D_1x1x1_sag_defaced.nii.gz). As with the equivalent images in the fMRI subdirectory, these have been "defaced" using the TractoR software package for anonymity.