TFMaps

Time-Frequency Analysis Toolbox

This folder contains MATLAB funcitons to perform time-frequency analysis of continuous data and compute power and coherence maps along a set of explanaotry variables. It includes two main functions: SetTFMapsParams for defining options such as which channels to use, which explanatory variables to use, how to bin them and over which subset of the data these maps should be computed. SetTFMapsParams then takes these options as an input, together with a structure containing your explanatory variables (called Nav here) and an array of responses for which you want to compute tuning curves in the spectral domain (e.g Lfp.Lfp_raw). See the Load function section for how the Nav and Lfp structures should be constructed.

SetTFMapsParams

The SetTFMapsParams function defines a set of parameters needed for time-frequency analysis of continuous data. It takes explanatory variables (Nav), response data (Lrep), and an optional input sampleTimes_Lrep (sampling times for signals in Lrep). It outputs a structure TFmapsparams containing parameters for the time-frequency analysis.

Parameters included in TFmapsparams:

• subset: Conditions and values used to subset the data in Nav.
• freqrange: Frequency range for time-frequency analysis.
• chidx: Subset of signals in Lrep considered for analysis.
• sampleRate: Sampling rate of the independent variables in Nav.
• sampleRate_raw: Sampling rate of the signals in Lrep.
• Spectrogram, Coherence: Flags to compute wavelet transforms and coherences.
• scalingFactor: Scaling factor on the response data for map computations.
• Xvariablename, Xbinedges, XsmthNbins: Parameters for mapping along X.
• Yvariablename, Ybinedges, YsmthNbins: Parameters for mapping along Y.
• occ_th: Occupancy threshold for map computations.
• nShuffle, kfold: Shuffle and cross-validation controls.
• parallel: Parallel computing option.

TFMapsAnalysis

The TFMapsAnalysis function performs time-frequency analysis and computes power and coherence maps based on the parameters defined in TFmapsparams. It takes as inputs a set of explanatory variables and experimental conditions (stored in Nav), response data (a ntimes x nchannels array, e.g. Lfp.Lfp_raw), a set of options to run the time-frequency analysis (stored in TFmapsparams), and sampleTimes_Lrep, the timestamps of samples in the response data array. This function returns tuning curves along the explanatory variables indicated in TFmapsparams for the power and the coherence across frequencies.

Fields included in the TFMaps structure:

• TFmapsparams: Parameters used for time-frequency analysis.
• wtMaps: Cell array of structures containing power maps across frequencies.
• wcMaps: Cell array of structures containing coherence maps across frequencies.
• fqbins: Frequency bins used for the analysis.

Usage Example

1. Load behavioral data and Lfp signals

   datapath = 'path/to/your/data';

   loadparams = SetLoadParams(datapath);

   Nav = LoaddataNav(loadparams);

   Lfp = LoaddataLfp(loadparams, Nav.sampleTimes);

   Lrep = Lfp.Lfp_raw;

2. Define time-frequency analysis parameters using SetTFMapsParams

   TFmapsparams = SetTFMapsParams(Nav, Lrep, Lfp.sampleTimes);

   adjust the paramters if needed. For instance:

   TFmapsparams.freqrange = [20 150];

   TFmapsparams.subset.Condition = [1 3 5];

   TFmapsparams.subset.Condition_op = 'ismember';

   TFmapsparams.subset.Spd = 2.5;

   TFmapsparams.subset.Spd = '>=';

   TFmapsparams.Xvariablename = 'Xpos';

   TFmapsparams.Xbinedges = 0:2:100;

   TFmapsparams.Xvariablename = [];

3. Perform time-frequency analysis using TFMapsAnalysis

   TFMaps = TFMapsAnalysis(Nav, Lrep, TFmapsparams, Lfp.sampleTimes);

For more detailed information and usage examples of each function, please refer to the function documentation, Tutorial4.1 and Tutorial_handson.

Developed by J. Fournier in August 2023 for the Summer school "Advanced computational analysis for behavioral and neurophysiological recordings."