Skip to main content

Package manager for NEURON

Project description

Build Status codecov Code style: black Documentation Status

Glia: NEURON package manager

Glia is an asset manager for NEURON. It collects mod files from different pip packages and compiles them into a central library that is automatically loaded into NEURON. This removes the need for compiling folder after folder with cluttered, duplicated mod files and allows you to focus on using these mechanisms across multiple models.

Packaging your mod files as a Glia package allows you to distribute them as dependencies of your Python models and delegates the installation, distribution, versioning and archiving of your assets to Python's packet manager pip.

To create Glia packages, check out the CLI tool Astrocyte. Astrocyte also allows you to organize your personal mod collection!

Usage

Glia can be installed from pip:

pip install nrn-glia

Glia will check whether packages have been added, changed or removed and will recompile and load the library if necessary. This means that except for importing Glia there's not much you need to do!

from neuron import h
import glia as g

section = h.Section(name="soma")
# Load your favourite Kv1 mechanism.
g.insert(section, "Kv1")

# Note: to load the library at import time you can import glia.library instead
import glia.library

Glia avoids conflicts between authors and even variants of the same mechanism and allows you to select sensible default preferences on many levels: globally, per script, per context or per function call.

Asset management

Glia allows for multiple assets to refer to the same mechanism by giving them a unique name per package. The standard naming convention is as follows:

glia__<package-name>__<asset-name>__<variant-name>

Double underscores in packages, assets or variant names are not allowed.

This naming convention allows for multiple people to provide an implementation of the same asset, and by using variants even one package can provide multiple variations on the same mechanism. The default variant is 0

If you install multiple packages that provide the same asset, or if you would like to specify another variant you will need to tell Glia which one you require. You can do so by setting your asset preferences.

Asset preferences

There are 4 different scopes for providing asset preferences:

  • Global scope: Selects a default mechanism asset everywhere.
  • Script scope: Selects a default mechanism asset for the remainder of the Python script.
  • Context scope: Select a preferred package or variant for all glia.insert calls within the context block.
  • Single use: Selects a mechanism asset for a single glia.insert call

Single use

Whenever you call glia.insert you can append your preferences for that insert:

g.insert('Kv1', pkg='not_my_models', variant='high_activity')

Context scope

Any glia.insert or glia.resolve call within the with statement will preferably use the given package or variant:

from patch import p
s = p.Section()
with g.context(pkg='not_my_models'):
  g.insert(s, 'Kv1')
  g.insert(s, 'Kv1', variant='high_activity')

You can also specify a dictionary with multiple asset-specific preferences:

from patch import p
s = p.Section()
with g.context(assets={
   'Kv1': {'package': 'not_my_models', 'variant': 'high_activity'},
   'HCN1': {'variant': 'revised'}
}):
  g.insert(s, 'Kv1')
  g.insert(s, 'HCN1')
  # Not affected by the context:
  g.insert(s, 'Kir2.3')

And you can even combine, preferring a certain package unless the dictionary specifies otherwise:

from patch import p
s = p.Section()
with g.context(assets={
   'Kv1': {'package': 'not_my_models', 'variant': 'high_activity'},
   'HCN1': {'variant': 'revised'}
}, package='some_pkg_name'):
  g.insert(s, 'Kv1')
  g.insert(s, 'HCN1')

Finally for those of you that have really crazy preferences you can even nest contexts, where the innermost preferences take priority.

Script scope

Use glia.select to select a preferred mechanism asset, similar to the single use syntax, for the remainder of the lifetime of the glia module:

section_global_Kv1 = h.Section()
section_local_Kv1 = h.Section()
g.insert(section_global_Kv1, 'Kv1') # Will use your global Kv1 mechanism
g.select('Kv1', pkg='not_my_models', variant='high_activity')
g.insert(section_local_Kv1, 'Kv1') # Will use the above selected Kv1 mechanism

Global scope

Applying global scope uses the Glia command-line tool and will configure glia to always select a mechanism asset as default.

Go to your favorite command-line and enter:

glia select Kv1 --pkg=some_pkg_name --variant=non_default

This will set your preference in any script you use.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distribution

nrn_glia-0.5.1-py3-none-any.whl (32.1 kB view details)

Uploaded Python 3

File details

Details for the file nrn_glia-0.5.1-py3-none-any.whl.

File metadata

  • Download URL: nrn_glia-0.5.1-py3-none-any.whl
  • Upload date:
  • Size: 32.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.7.1 importlib_metadata/4.8.2 pkginfo/1.8.2 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.62.3 CPython/3.9.9

File hashes

Hashes for nrn_glia-0.5.1-py3-none-any.whl
Algorithm Hash digest
SHA256 9b4782b6166bcfe9aced77a060a06f4b6f688804a45c58751a42d9fb2294f3b4
MD5 f25860023b60d9774aeafc952a750086
BLAKE2b-256 c791a76a3cae9635f6bb1e2cc4d5199e4e4553f83c83479b12a193270c0ab497

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page