These Matlab scripts provide n:m phase-locking analyses used in Scheffer-Teixeira & Tort, eLife 2016 (https://doi.org/10.7554/eLife.20515). To reproduce our results, we made the analyzed dataset available at http://dx.doi.org/10.5061/dryad.12t21
Brief instructions:
-
Open the caller script.
-
Run the first cell and select the folder where this package is to be stored.
-
In the second cell you can choose which signal type you want to analyze. The default signal is a white noise. If you want to use real LFP, uncomment the line containing 'real_lfp'.
-
Choose which surrogates you want to include. The default uses all analyses. Just comment out those you are not interested.
-
Define the parameters of your analysis. Each parameter is described in the caller routine.
When using real signals, a window will open and ask for a .mat file. You can actually choose more than one file, which can be sessions from the same animal, several sessions from several animals, or just one session from one animal. The only requisite is that the file must contain an array structured as channels x time which should be named "LFP". As a template, we refer to our data avaiable at Dryad.
To avoid memory out messages, the script opens and runs each file in sequential order.
After plotting the results, all analyses are stored in a variable called Data.
In addition to the toolbox files, we also provide separated .m files for simulating and analyzing Kuramoto oscillators; these files reproduce Figures 1 and 3 of our paper and are named accordingly.