elephant.spade.pvalue_spectrum

elephant.spade.pvalue_spectrum(spiketrains, bin_size, winlen, dither, n_surr, min_spikes=2, min_occ=2, max_spikes=None, max_occ=None, min_neu=1, spectrum='#', surr_method='dither_spikes', **surr_kwargs)

Compute the p-value spectrum of pattern signatures extracted from surrogates of parallel spike trains, under the null hypothesis of independent spiking.

• n_surr surrogates are obtained from each spike train by spike dithering

• pattern candidates (concepts) are collected from each surrogate data

• the signatures (number of spikes, number of occurrences) of all patterns are computed, and their occurrence probability estimated by their occurrence frequency (p-value spectrum)

Parameters:

spiketrainslist of neo.core.SpikeTrain

List containing the parallel spike trains to analyze

bin_sizepq.Quantity

The time precision used to discretize the spiketrains (binning).

winlenint

The size (number of bins) of the sliding window used for the analysis. The maximal length of a pattern (delay between first and last spike) is then given by winlen*bin_size

ditherpq.Quantity

Amount of spike time dithering for creating the surrogates for filtering the pattern spectrum. A spike at time t is placed randomly within [t-dither, t+dither] (see also elephant.spike_train_surrogates.dither_spikes()). Default: 15*pq.s

n_surrint

Number of surrogates to generate to compute the p-value spectrum. This number should be large (n_surr>=1000 is recommended for 100 spike trains in spiketrains). If n_surr is 0, then the p-value spectrum is not computed. Default: 0

min_spikesint, optional

Minimum number of spikes of a sequence to be considered a pattern. Default: 2

min_occint, optional

Minimum number of occurrences of a sequence to be considered as a pattern. Default: 2

max_spikesint, optional

Maximum number of spikes of a sequence to be considered a pattern. If None no maximal number of spikes is considered. Default: None

max_occint, optional

Maximum number of occurrences of a sequence to be considered as a pattern. If None, no maximal number of occurrences is considered. Default: None

min_neuint, optional

Minimum number of neurons in a sequence to be considered a pattern. Default: 1

spectrum{‘#’, ‘3d#’}, optional

Defines the signature of the patterns.

‘#’: pattern spectrum using the as signature the pair:

(number of spikes, number of occurrence)

‘3d#’: pattern spectrum using the as signature the triplets:

(number of spikes, number of occurrence, difference between last and first spike of the pattern)

Default: ‘#’

surr_methodstr

Method that is used to generate the surrogates. You can use every method defined in elephant.spike_train_surrogates.dither_spikes(). Default: ‘dither_spikes’

surr_kwargs

Keyword arguments that are passed to the surrogate methods.

Returns:

pv_speclist

if spectrum == ‘#’:

A list of triplets (z,c,p), where (z,c) is a pattern signature and p is the corresponding p-value (fraction of surrogates containing signatures (z*,c*)>=(z,c)).

if spectrum == ‘3d#’:

A list of triplets (z,c,l,p), where (z,c,l) is a pattern signature and p is the corresponding p-value (fraction of surrogates containing signatures (z*,c*,l*)>=(z,c,l)).

Signatures whose empirical p-value is 0 are not listed.