A toolbox for detecting/quantifying/comparing reactivation and replay in neural data. Some of these methods and tests were described in Tingley & Peyrache (2019)
- Clone this repo to your local machine using
https://github.com/davidtingley/RnR_methods
- Run
addpath(genpath(pwd))
from the \RnR_methods directory
ReactStrength
, to compute reactivation (i.e. 0-lag neuronal correlation) with PCA or ICA. Type 'help ReactStrength' for mor info
replay_Bayesian
uses the Bayesian method (Zhang et al., 1998) to quantify replay scores when given an average firing rate template and a set of candidate replay events. It returns both the maximum integral under the line of best fit, using the Radon transform (Davidson et al., 2009) and the linear weighted correlation of the posterior probability matrix (Wu & Foster 2014).replay_RankOrder
uses the rank order correlation (Foster & Wilson 2006) to quantify replay scores when given an average firing rate template and a set of candidate replay events.
compareReplayMethods
, to compare different reactivation methods (Fig. 3 in Tingley&Peyrache)compareNoiseDegradation
, to compare how noise in data affects replay (and reactivation) (Fig. 4 in Tingley&Peyrache)rankOrder_falsePositive_test
, to see how limited length sequences and within even shuffling can lead to high FP rates/compareBinSize
, to compare how bin size affects replay detection