Scalp-Negative Medial Temporal Interictal Epileptic Discharges Alter Large-Scale Brain Networks: A Simultaneous High-Density EEG And Intracranial EEG Study

List of files containing the connectivity matrices, network measures, and results of the statistical tests from the manuscript: Scalp-Negative Medial Temporal Interictal Epileptic Discharges Alter Large-Scale Brain Networks: A Simultaneous High-Density EEG And Intracranial EEG Study by N. Roehri, P. De Stefano, L. Spinelli, R. Marquis, S. Lagarde, S. Momjian, M. Seeck, S. Vulliemoz from the EEG and Epilepsy Unit, University Hospitals and Faculty of Medicine, University of Geneva, Rue Gabrielle-Perret-Gentil 4, 1205 Geneva, SwitzerlandGeneral description:*.tsv is a simple format with 'Tab-separated values'; it can be opened in any notepad software and Excel.*.mat is a Matlab file; it can be opened with Matalb, Python or R.List of files: <br>sub-##_connectivity.mat:<br><br>Matlab files containing the connectivity matrices obtained with the weighted phase lag index of subject ## (i.e. 01 to 09) for each frequency band and each condition. The structure 'results' contains different fields and subfields:<br>Three modality fields:<br>.eeg: field containing the results for the EEG,<br>.ieeg: field containing the results for the iEEG at the channel level,<br>.ieegROI: field containing the results for the iEEG at the ROI level,<br><br>Other fields:<br>.tap_freq: tappering window used,<br>.fbands: lower and upper bounds of the frequency bands of interest,<br>.fband_str: name of the above frequency bands.<br><br>Each modality field is structured the following way:<br>.labels: the label of the ROIs (for eeg and ieegROI) or the channel (for ieeg),<br>.freq: the value of the frequency bins for the frequency band of interest.<br>.noIED and .scalpnegIED fields contain each a field for each frequency band (e.g., .delta) with:<br>.wpli: a NxN matrix with the connectivity values obtained with the weighted phase lag index (wpli) of the frequency band of interest,<br>.spectr: a Nx1 vector with the spectral power of the frequency band of interest,<br>.itc: a Nx1 vector with the spectral inter-trial coherence of the frequency band of interest,.labels: the label of the ROIs (for eeg and ieegROI) or the channel (for ieeg),<br>.freq: the value of the frequency bins for the frequency band of interest.<br>N is either the number of ROIs or the number of ieeg channels.Whole brain connectivity analysisglobal_network_features.mat:<br><br>Matlab structure containing the values of the whole brain network integration and segregation for the ROI traces derived from hdEEG (hdEEG(ROI)), the iEEG traces (iEEG(channel)) and the iEEG summarised at the ROI level (iEEG(ROI)) for the two conditions and each frequency band.<br>The structure follows this organisation:<br>the first level is the metric (i.e. .integration or .segregation) followed by the modality (i.e., .eeg, .ieeg, .ieegROI), and finally the condition (i.e., .noIED or .scalpnegIED) containing a 9x4 matrix, the row correspond to the patients, the column to the frequency band (i.e. delta, theta, alpha, beta).<br>integration_segregation_wpli.tsv<br>Table giving the result of the the Wilcoxon signed rank test comparing the whole brain network features of the two conditions for different modality and frequency bands.<br>'network_metric' gives the network metric (i.e. integration or segregation);<br>'modality' gives the modality of the analysed traces and its level, channel or ROI for iEEG (i.e., hdEEG_ROI, iEEG_channel, iEEG_ROI);<br>'freq_band' the frequency band under investigation;<br>'median_diff' is the median of the difference between the conditions across patients;<br>'N_patient_with_increase' is the number of patients for which the network feature is higher for the scalp negative IEDs than for the no IED;<br>'p_value' and 'adjusted_p_value' are the p-values and FDR-adjusted p-values of the Wilcoxon signed rank test.<br>Time-frequency analysisieeg_ersp_posclusters.tsv<br>Table containing the results of the cluster based permutation testing comparing scalp negative IEDs to no IED using iEEG and event-related spectral perturbation (ERSP) as a metric.<br>The column 'prob' gives the p-value of the cluster, 'clusterstat' gives the summed-t quantifying the strength of the cluster, 'stddev' the standard deviation of the probability estimation and 'cirange' the confidence range of the probability estimation.<br>ieeg_itc_posclusters.tsv<br>Table containing the results of the cluster based permutation testing comparing scalp negative IEDs to no IED using iEEG and inter-trial coherence (ITC) as a metric.<br>The columns are the same as in ieeg_ersp_posclusters.tsv<br>roi_ersp_posclusters.tsv<br>Table containing the results of the cluster based permutation testing comparing scalp negative IEDs to no IED using the ROI traces obtained from the hdEEG and event-related spectral perturbation (ERSP) as a metric.<br>The columns are the same as in ieeg_ersp_posclusters.tsv<br>roi_itc_posclusters.tsv<br>Table containing the results of the cluster based permutation testing comparing scalp negative IEDs to no IED using the ROI traces obtained from the hdEEG and inter-trial coherence (ITC) as a metric.<br>The columns are the same as in ieeg_ersp_posclusters.tsvROI analysis<br>segregation_ieeg_roi_wpli.tsv<br>Table giving the result of the the Wilcoxon signed rank test comparing the ROI-level segregation of the two conditions for the iEEG(ROI) and frequency bands.<br>'freq_band' represents the frequency band under investigation;<br>'ROI_name' gives the name of the ROI under investigation,<br>'median_diff' is the median of the difference between the conditions across patients;<br>'N_patient_with_increase' is the number of patients for which the network feature of that ROI is higher for the scalp negative IEDs than for the no IED;<br>'p_value' and 'adjusted_p_value' are the p-values and FDR-adjusted p-values of the Wilcoxon signed rank test.<br>integration_ieeg_roi_wpli.tsv<br>Table giving the result of the the Wilcoxon signed rank test comparing the ROI-level integration of the two conditions for the iEEG(ROI) and frequency bands.<br>The columns follow those of segregation_ieeg_roi_wpli.tsv<br>segregation_roi_wpli.tsv<br>Table giving the result of the the Wilcoxon signed rank test comparing the ROI-level segregation of the two conditions for the hdEEG(ROI) and frequency bands.<br>The columns follow those of segregation_ieeg_roi_wpli.tsv<br>integration_roi_wpli.tsv<br>Table giving the result of the the Wilcoxon signed rank test comparing the ROI-level integration of the two conditions for the hdEEG(ROI) and frequency bands.<br>The columns follow those of segregation_ieeg_roi_wpli.tsv