Skip to content

dbbs-lab/patch

Repository files navigation

build codecov Code style: black Documentation Status

Drop in replacement

*No ducks were punched during the construction of this monkey patch.

Installation

pip install nrn-patch

Minimal requirements

  • Python 3.8+
  • NEURON 8.0+

Philosophy

Pythonic reinvention of the NEURON Python interface:

  • Drop-in replacement: No breaking changes between Patch and NEURON. All your code works as is. Patch automagically fixes bugs and reduces complexity in NEURON by having an opinion, and adds new convenient methods.
  • Loud errors: silent failures and gotchas are caught and raise loud errors; so that when it runs, it runs.
  • Strong referencing: Objects connected to each other will never disappear until you disconnect them.
  • Demystified: The simplest form of each instruction does the most obvious thing. Patch frequently replaces 5 or more undocumented mystical voodoo steps by 1 clearly named function.
  • Just add water: Serial or parallel, doesn't matter, p.run() will run your simulation with no extra steps. No load_file, finitialize, fadvance, run, continuerun, tstop, set_maxstep, setup_transfer, setgid2node, gid_connect, cell, outputcell, psolve required.

Basic usage

Use it like you would use NEURON. The wrapper doesn't make any changes to the interface, it just patches up some of the more frequent and ridiculous gotchas.

Patch supplies a new interpreter p, the PythonHocInterpreter which wraps the standard HOC interpreter h provided by NEURON. Any objects returned will either be PythonHocObject's wrapping their corresponding NEURON object, or whatever NEURON returns.

When using just Patch the difference between NEURON and Patch objects is handled transparently, but if you wish to mix interpreters you can transform all Patch objects back to NEURON objects with obj.__neuron__(), or the transform function.

Example

from patch import p
import glia as g

section = p.Section()
point_process = g.insert(section, "AMPA")
point_process.stimulate(start=10, number=5, interval=10, weight=0.04)

.stimulate creates both the NetCon and NetStim, and sets their properties, and stores references to eachother so that Python doesn't garbage collect anything until we're done with it.

Even when you forget to set values, the Patch defaults have a visible effect. Here we forget to set the duration of a voltage clamp, but by default it will be active the first 100ms of the simulation, so that you can see that the voltage clamp is in fact inserted and working, but you forgot to set the duration:

section.vclamp(holding=-70, voltage=20)

About

A Pythonic object-oriented drop-in replacement for the Python interface to NEURON

Resources

License

Stars

Watchers

Forks

Packages

No packages published