File Exchange

image thumbnail


version (876 KB) by Thomas Deneux
Spike extraction from calcium signals (Deneux et al., 2016)

1 Download

Updated 19 Feb 2018

GitHub view license on GitHub

MLspike is an algorithm to reconstruct neuronal spiking activity from noisy calcium recordings. Its description and benchmarking can be found in (Deneux et al. 2016).

In addition to the help of individual functions, help can be found in the Help browser: type `doc`, then go to 'Supplemental Software' > 'MLspike toolbox'. Note the demo scripts spk_demoGUI and spk_demo to get accustomed with the main functions.

MLspike requires the Brick toolbox (

Comments and Ratings (1)

Hi - code looks interesting but running the spk_demo.m fails on my system. The issues start in line 133 (first call of spk_autocalibration). Please advise. Full output copy-pasted below:
Trial>> spk_demo

a =


tau =


sigma =


tps_mlspike 6/6
estimated sigma: 0.021
Undefined function or variable 'MM'.

Error in tps_mlspikes>backward_driftstate (line 896)
MM = cat(1,MM{:});

Error in tps_mlspikes>backward (line 435)
[n Ffit par.F0 LL xest] = backward_driftstate(F,par);

Error in tps_mlspikes (line 131)
[varargout{:}] = backward(F,par);

Error in spk_autocalibration>autocalibration (line 182)
[n{i} fit{i} dum dum dum drift{i}] = tps_mlspikes(calcium{i},par); %#ok<ASGLU>

Error in spk_autocalibration (line 53)
[tau amp sigmaest events] = autocalibration(calcium,pax,sigmaonly);

Error in spk_demo (line 133)
[tauest aest sigmaest] = spk_autocalibration(calcium,pax)




added image and reference to brick toolbox

MATLAB Release Compatibility
Created with R2013a
Compatible with any release
Platform Compatibility
Windows macOS Linux