www.neuron.yale.edu

I'm trying to figure out Neuron's code, and have gotten totally lost.

If I look at the code for fadvance, I see that it's repeatedly calling nrn_solve, advancing the simulation through time. Within nrn_solve, there are triang and bksub routines that do the Gaussian elimination over the tree of compartments. These work on a node structure that has members _a, _b, _d, and _rhs, but I can't find any place where these are initialized or solved for.

It seems to me that there HAS to be somewhere that this is done: where the variables are set to something derived from compartment voltage differences, membrane capacitance, resistance, and so on, but where? And how? The code, and all the papers & references &'ve been able to find, seem to assume a whole body of pre-existing knowledge that I don't have, and don't know where to find. Can someone give me a clue?

Thanks,
James

See
Details of fadvance()
in chapter 7 of the NEURON book (get draft here
http://www.neuron.yale.edu/ftp/ted/book ... xedref.pdf
). Note that fadvance sets up the equation matrix by calling treeset.c

Thanks, that's useful. The chapter title and the first couple of pages are rather misleading, though. I looked at them a couple of weeks ago, saw the picture and the word "GUI", and passed it by.