 |
Jon Newman PhD Candidate |
 |
|
|
Jon Newman PhD Candidate |
|
NeuroRighter (NR) is a free, MS Windows-based data recording and stimulation system for mulitelectrode arrays created by John Rolston during his thesis work at Emory and Georgia Tech. It has all the bells and whistles of an extracellular recording system (probably more than usual, actually) and its source code is available to all for modification and customization. Additionally NR is inexpensive to create (less then USD 10,000, including computer), runs on open-source (GPLv3 protected) software written in C#, is easily modifiable (in terms of hardware and software) and well documented. John has gone on finish the MD portion of his MD/PhD and has left me in charge of maintaining NR in his absence. For detailed documentation on the system, check out the following goodies:
The NeuroRighter site - actively updated and maintained user manual packed with up to date info on the system.
The NeuroRighter user's group - post questions on this forum to get answers from other NR users.
John Rolston's site - the man and the myth...
National Instruments Developer Zone - NeuroRighter uses NI DAQ's and software drivers. NI has fantastic user forums for help in modifying NR for your own experiments.
SqueakySpk is an open-source (GPLv3 protected) matlab class designed for basic multichannel, extracellular data analysis. One can use a SqueakySpk object to store extracellular data with along with methods used for artifact/noise rejection and spike sorting. By storing data as a SqueakySpk object:
To install SqueakySpk, download from the link below and follow the instructions provided in ReadMe.txt
Version |
Changes | Date |
- First Release |
2010-06-22 |
If you wish to contribute to SqueakySpk's development, then email me for write permission to the Google Code SVN repository. Even without write permission you are free to download and modify the code from the SVN Repo, but you wont' be able to commit any changes (which would be a shame because they may be beneficial to others).
During the spring of 2009, I carried out a semester long project attempting to control chaotic bursting in cortical networks grown on MEA's. Chaotic attractors are distinguished by the existence of an infinity of unstable periodic orbits embedded in the limit set of the system. These orbits are dense in the attractor, and therefore local dynamics can be well approximated by them. To control chaos locally, one simply forms a linear approximate model of the system about one of these orbits (fixed points of the discrete time system resulting from the introduction of a Poincaré section) and apply the well know pole placement technique from linear controls theory. The caveat is that this requires the extraction of fixed points from time series data to control a physical system for which an accurate model can not be produced. FPTools is a matlab package that implements Paul So's transform to concentrate time series data about fixed points, test the significance of reported fixed points using statistics of extremes and form a linear approximate model about significant fixed points. This package includes documentation for installation and use as well as several examples.
fptools version 0.6 [matlab]
A while ago, I implemented some multi-point shooting routines in Mathematica that can be used to find unstable cycles in chaotic attractors to arbitrary length if a good knowledge of the symbolic dynamics is known. This code is implemented for the Rössler flow, but should be useful as an example of the method.
MPSM for Rossler [mathematica]
examples of MPSM for Rossler [mathematica]
cycle visualization [mathematica]
In the zip file below, you will find all the models that were employed for my study on multirhythmic bursting. The paper that made use of these scripts, entitled Mechanism, Dynamics and Biological Existence of Multistability in a Large Class of Bursting Neurons, can be found in the publications section. Most of these routines rely on some finicky numerical integration methods, so don't be too surprised if you need to mess with initial conditions and/or the type of solver to get convergence.
multiryhthmic bursting scripts [matlab, XPP]
| Copyright (c) 2010, Jon Newman | |