Getting started. Again.

Post Reply
plasticity
Posts: 17
Joined: Fri Oct 19, 2018 3:09 pm

Getting started. Again.

Post by plasticity »

I introduced myself to NEURON a couple years ago before getting distracted by more macroscopic neuroscience questions. Now I am back to increasing my understanding of the microscopic. I am back to NEURON.

I am a physicist and I have done previous reading about the cable equation, the HH equations and simple models of neuron behavior. Here’s my present stumbling block: I do not understand what the scope of NEURON is and I do not understand which mathematical models underpin that scope. My question: How do I fix these gaps in my understanding? I would like to make significant progress in fixing those gaps in my knowledge before taking the NEURON course.
ted
Site Admin
Posts: 6287
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Getting started. Again.

Post by ted »

Welcome, physicist. The center of NEURON's domain of utility is mechanistic modeling of biological neurons and networks in which the complex anatomical and biophysical properties of cells (including but not limited to ion channels) and the interconnections between them are important. NEURON models can also involve point neurons such as single compartment reductions of biophysical model cells and dynamical abstractions including Izhikevich, Morris-Lecar, adex, and garden variety integrate and fire "cells". The anatomical scale of a NEURON model can range from representations of intra- and/or extracellular reaction-diffusion, subcellular structures (individual neurites, spines, cytoplasmic organelles), to very large scale network models (see especially the European Human Brain Project). Simulations can be executed on individual processors or can take advantage of multithreaded parallelization and/or MPI-dependent distributed execution.

With regard to equations, the cable equation is of course relevant to any model cell that has significant electrical extent, but NEURON shields you from having to deal with the cable equation directly. ODEs (e.g. Hodgkin-Huxley-style channel descriptions) and state transition models (Markov models) are relevant to many phenomena that are important to neuronal function, e.g. channel gating, ligand-receptor interactions. NEURON's Channel Builder and NMODL model specification language simplify the task of adding new ion channels, synaptic mechanisms, and drug-receptor interactions to NEURON.

That's not a complete description, but it should give you an idea of what NEURON can deal with.

If you're looking for stuff to read, you might try the first few chapters of The NEURON Book; if you don't have the book, you might want to examine some of the preprints from https://www.neuron.yale.edu/ftp/ted/book/revisions/ All of the book's coding examples are in hoc, but the concepts introduced are still valid. We advise that all new modeling projects use Python instead of hoc because of Python's many modern and powerful features (especially introspection); you'll find a tutorial on using Python to build and work with NEURON models at https://neuron.yale.edu/neuron/docs/scr ... ron-basics
plasticity
Posts: 17
Joined: Fri Oct 19, 2018 3:09 pm

Re: Getting started. Again.

Post by plasticity »

Thank you.

I have read the first few chapters of The Neuron Book. I don’t like to use black box code so how do I learn more about the mathematics and numerical calculations underlying each of the components used to build models with NEURON?
ted
Site Admin
Posts: 6287
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Getting started. Again.

Post by ted »

WRT NEURON, feel free to keep reading through at least chapter 9. Supplement that with the articles listed at https://neuron.yale.edu/neuron/publications/biblio (that list needs to be updated to include articles from the past 3 years, but it already contains enough papers to keep you busy for a while). And by all means, download the source code and read that.

In the meantime, try not to use any OS (especially not Windows or MacOS or whatever Apple calls it currently), web browser, email client, text editor, programming language etc. without first checking its source code.
plasticity
Posts: 17
Joined: Fri Oct 19, 2018 3:09 pm

Re: Getting started. Again.

Post by plasticity »

Ted

I am looking for references, preferably a book, which covers the models used in NEURON and the mathematics (and physics) underlying those models. I am presently not particularly interested in the source code that implements the underlying mathematics.

PS: Pardon the delay. It took longer than I thought it would to code a new OS.
ted
Site Admin
Posts: 6287
Joined: Wed May 18, 2005 4:50 pm
Location: Yale University School of Medicine
Contact:

Re: Getting started. Again.

Post by ted »

the models used in NEURON
I'm not quite sure what you mean. If you mean "how are the properties of biological neurons represented in mechanistic models of neurons implemented with NEURON" then chapters 3-5 of The NEURON Book are for you. What isn't covered in those chapters? Drug-receptor interactions, diffusion, stochasticity of channel gating and transmitter release. You might find some of that in Johnston and Wu's Fundamentals of Cellular Neurophysiology, or in Hammond's Cellular and Molecular Neurophysiology. You might also find it useful to look for articles by Robert McDougal about reaction diffusion. There aren't a lot of good books on these subjects; most are merely collections of articles, and some are either dumbed way down or seem designed to make obscure that which others have presented clearly elsewhere.
plasticity
Posts: 17
Joined: Fri Oct 19, 2018 3:09 pm

Re: Getting started. Again.

Post by plasticity »

Thanks

I will read the materials and get back to you if I need more before taking the class.
Post Reply