Supporting data and scripts for the paper "Variability in the location
of High Frequency Oscillations during prolonged intracranial EEG
recordings"
This compilation of data and scripts are to support the following
publication:
Stephen V. Gliske, Zachary T. Irwin, Cynthia Chestek, Garnett
L. Hegeman, Benjamin Brinkmann, Oren Sagher, Hugh J. L. Garton, Greg
A. Worrell, William C. Stacey. "Variability in the location of High
Frequency Oscillations during prolonged intracranial EEG recordings."
Nature Communications (in press).
More detail about the motivation, methods, and data set can be found
therein. The main input data for these scripts are found in the "data"
folder, and consist of the times of HFO detections for three different
HFO detectors and for four cohorts (UM is treated as two cohorts: one
which uses all NREM data and the other which uses 1-3 AM data).
The main point of entry for the matlab scripts are "analyze_data.m"
and "all_plots.m", which cover the main analysis and plots,
respectively. The analysis script takes approximately two hours to
run over all detectors and cohorts. It computes the blind source
separation for all the rates. That procedure is stochastic, and thus
results may vary slightly from the paper. The plotting script shows
examples of how to make the plots corresponding to the figures in the
paper.
The other scripts are as follows:
addLetter.m -- adds subplot letters
blind_source_separation.m -- performs blind source separation, called by
analyze_data.m
compute_asym.m -- computes asymmetries, called by plot_asyms.m
determineOrder.m -- ancillary script for plot_factorization.m
gbar.m -- ancillary script for plot_max_chan.m
gbar2.m -- ancillary script for plot_categories.m
makeRateHeatMap.m -- ancillary script for plot_factorization.m
plot_asyms.m -- makes the asymmetry plots
plot_categories.m -- makes the plot of categories of temporal
variability
plot_factorization.m -- makes the plots of rate and the blind source
separation
plot_heatMap_patches.m -- ancillary script for plot_UMHS_0007.m
plot_max_chan.m -- makes the plot for Figure A5
plot_scan_UM_rec_time.m -- makes the plot of scanning over how much
recording time is used to determine category
plot_UMHS_0007.m -- makes plot of subject UMHS-0007's HFO rate
versus seizure times (electrographic and
clinical)
robust_nnmf.m -- the non-negative matrix factorization
portion of the blind source separation
Additionally, there is a data file with times of seizures for subject
UMHS-0007 titled "UMHS-0007-electrographic-seizures.txt" and an empty
directory "plots" where the plots are saved.
Created in 2018 by S. Gliske (sgliske@umich.edu)
Released under the CC-BY-NC-4.0 License
http://creativecommons.org/licenses/by-nc/4.0/