There are many possible analyses of electrophysiology data, so it depends largely on what kinds of analyses you plan on doing, what your hypotheses are, etc. Here's a resource that might help: amazon.com/Analyzing-Neural-Time-Series-Data/dp/0262019876/

Which types of electrophysiological analysis you want to learn? For patch clamp electrophysiological data analysis you can read several online tutorials from HEKA electronics, Axon clamp etc. Also, for analysis of data you should have an outline that what you want to analyse like if you want to learn analysis of current -voltage relationship for any ion channel then you can go to to the document attached with this mail.I am not sure about your work although there are several way to learn but cab be tell after knowing your objectives.

Thanks for both of the responses. I am doing in vivo electrophysiology--I have rats performing in a behavioral task while recording (bundles of electrodes targeting one specific brain region). I am looking for references that can give me an idea of the best strategy to cluster cut, correlate cells on different channels and then align spike data with timestamped events from the behavior... I have not been able to find a great reference, so if anyone has suggestions, that would be helpful!

"Observed Brain Dynamics" by Partha Mitra is a fantastically written book [1]. Simply a delight to read. It has a good chapter on spike sorting, which seems to be you area of most interest. The major strength of this book is spectral estimation and and spike-field analyses. Chronux 2.1 is a free Matlab toolbox that puts this book into action. This book stemmed out of a course he taught at Woods Hole on the subject. http://hermes.mbl.edu/education/courses/special_topics/neufo.html

If you simply want a good clustering algorithm, check out Quiroga' superparamagnetic algorithm [2]. KlustaKwik is another good option [3]. Besides these two options, Larry Carin's team has probably the best algorithm out there [4], but I don't think they have released their code just yet.

[1] Mitra P, Bokil H. Observed brain dynamics. Oxford University Press, New York, New York, USA: New York, New York, USA, 2008.

[2] Quiroga RQ, Nadasdy Z, Ben-Shaul Y. Unsupervised Spike Detection and Sorting with Wavelets and Superparamagnetic Clustering. Neural Computation 2004; 16: 1661–87.

[3] Portet F, Hernández AI, Carrault DG. Evaluation of real-time QRS detection algorithms in variable contexts. Med Biol Eng Comput 2005; 43: 379–85.

[4] Carlson DE, Vogelstein JT, Wu Q, Lian W, Zhou M, Stoetzner CR, et al. Multichannel Electrophysiological Spike Sorting via Joint Dictionary Learning and Mixture Modeling. IEEE Transactions on Biomedical Engineering 2014; 61: 41–54.

I like to give my students this paper as a starting point for recording:

Chronic Recording of Extracellular Neuronal Activity in Behaving Animals

Ronald Szymusiak1, Douglas Nitz2

1 University of California, Los Angeles, California, 2 Neurosciences Institute, La Jolla, California

Publication Name: Current Protocols in Neuroscience

Unit Number: Unit 6.16

DOI: 10.1002/0471142301.ns0616s21

Online Posting Date: February, 2003

The Buzsaki Lab has open source software posted if you need those resources (including Ken Harris' KlustaKwik mentioned above). If you already have software, like Offline Sorter - I recommend visiting a lab (or sitting down with someone in your own lab) where they are expert at cluster cutting and getting some personal tutoring on the process. If you test almost all the auto algorithms, depending on your brain region and which algorithm - it still cuts noise as cells occasionally. Additional manual refinement of sorting allows you to fine-tune the clusters. You can also learn to take notes and draw as you record and then look for the clusters along all dimensions afterwards to tune-up your recognition. If you have a huge number of channels, you might want to sort automatically - as this will not be feasible.

All of the above information is great!! Thanks for everyone that responded!

Dear Caitlin,

my colleague developed for our purposes a matlab-toolbox which offers functions to analyze spike data.

"MLIB contains functions for a) assessing spike sorting quality / unit isolation, and b) constructing all sorts of peri-stimulus time histograms as well as raster displays and spike density functions constructed with various filter kernels."

Perhaps this could be helpful for you, once you have the data and know what you would like to analyze...

...

Data MLIB Toolbox Version 5