CoCoNAD (Continuous-time Closed Neuron Assembly Detection) is a method to find frequent synchronous joint events in parallel point processes, which has applications in the analysis of parallel spike trains. The idea is to provide a method to test the temporal coincidence coding hypothesis, that is, that stimuli are encoded by temporally coincident spiking of groups of neurons, sometimes called cell assemblies.

Languages used C, Matlab
Supported OSs 64-bit Linux
Download coconad-m-0.9.1.tar.gz (2016-02-12)

coconad.m continuous-time closed neuron assembly detection
estpsp.m estimate a pattern spectrum from original data
genpsp.m generate a pattern spectrum from surrogate data
patred.m pattern set reduction based on a preference relation
pats2file.m export patterns to text file
psp2bdr.m extract a decision border from a pattern spectrum
readTrains.m read spike trains from file

CoCoNAD is also available for Python (PyCoCo), R (CoCo4R), Java (JNICoCo), and as a command line program. There is also a GUI based on CoCoNAD for Java (CoCoGUI).


Picado-Muiño D, Borgelt C (2014). Frequent item set mining for sequential data: synchrony in neuronal spike trains. Intelligent Data Analysis 18(6):997-1012.

Borgelt C, Picado-Muiño D (2013). Finding frequent patterns in parallel point processes. Proc. 12th Int. Symposium on Intelligent Data Analysis, 116-126.

Picado-Muiño D, Borgelt C, Berger D, Gerstein G, Grün S (2013). Finding neural assemblies with frequent item set mining. Frontiers in Neuroinformatics 7:9.

Torre E, Picado-Muiño D, Denker M, Borgelt C, Grün S (2013). Statistical evaluation of synchronous spike patterns extracted by frequent itemset mining. Frontiers in Computational Neuroscience, 7:132.

Borgelt C, Picado-Muiño D (2014). Simple pattern spectrum estimation for fast pattern filtering with CoCoNAD. Proc. 13th Int. Symposium on Intelligent Data Analysis, 37-48.