A single package in Python unifying scripts and modules for reading, writing, simulating and analysing NeuroML2/LEMS models.
Builds on: libNeuroML & PyLEMS and wraps functionality from jNeuroML.
pyNeuroML relies on additional software to carry out its functions:
- Java Runtime environment (JRE)
- dot (from Graphviz)
- lxml
On most Linux systems, these can be installed using the default package manager. On Ubuntu based distributions:
sudo apt-get install python-lxml graphviz openjdk-11-jdk
pyNeuroML can be installed with pip (preferably in a virtual environment):
pip install pyneuroml
A number of extra packages are also provided for convenience. You can install these to pull in other dependencies if required:
pip install pyneuroml[neuron] # for NEURON simulation backend
pip install pyneuroml[brian] # for Brian2 simulation backend
pip install pyneuroml[netpyne] # for NetPyNE simulation backend
pip install pyneuroml[povray] # for povray functions
pip install pyneuroml[hdf5] # for HDF5 support
pip install pyneuroml[analysis] # for analysis functions
pip install pyneuroml[tune] # for tuning/fitting functions
pip install pyneuroml[vispy] # for 3D interactive morphology plotting using vispy
pip install pyneuroml[plotly] # for interactive plotting with plotly
pip install pyneuroml[nsg] # pulls in pynsgr to use NSG
pip install pyneuroml[combine] # includes libsbml, libsedml
pip install pyneuroml[tellurium] # for Tellurium simulation backend
pip install pyneuroml[all] # installs all of the above
pip install pyneuroml[dev] # installs all of the above and other test related packages
pip install pyneuroml[doc] # for building documentation
Please see the setup.cfg
file for more details.
The NeuroFedora community initiative provides pyNeuroML for use on the Fedora Linux Distribution. Fedora users can install pyNeuroML using the following commands:
sudo dnf copr enable @neurofedora/neurofedora-extra
sudo dnf install python3-pyneuroml
This will also pull in all the necessary dependencies (Java, lxml, graphviz). Please see the project documentation for more information.
Clone the repository:
git clone https://github.com/NeuroML/pyNeuroML.git
cd pyNeuroML
It should be possible to install pyNeuroML using just:
pip install .
To develop pyNeuroML, you can use the dev
extra and the development
branch:
git clone https://github.com/NeuroML/pyNeuroML.git
cd pyNeuroML
git checkout development
pip install .[dev]
1) Single Python package for NeuroML2/LEMS
One Python package which can be installed using pip & a user has everything they need to work with NeuroML2/LEMS files:
- libNeuroML
- PyLEMS
- A bundled version of jNeuroML which can be used to run simulations
2) Run models using jNeuroML or PyLEMS
Ability to run NeuroML2/LEMS models using jLEMS/jNeuroML (with bundled jar) or PyLEMS (todo...)
Uses similar command line interface to jNeuroML, i.e. based on jnml
Try:
pynml -h
to list current options.
3) Access to export & import options of jNeuroML
All export & import options of jNeuroML available through easy command line interface (see here for progress) & through Python methods.
Example of export of NeuroML2/LEMS to NEURON and execution of generated code using single method is here.
4) Helper Python scripts
Lots of helper scripts for commonly used functions, e.g. generating a firing frequency vs injected current plot, generating a LEMS file for use with a NeuroML2 file,
5) Analysis of ion channels
Generation of plots of activation rates for ion channels from NeuroML2 channel file (example):
pynml-channelanalysis NaConductance.channel.nml
Generation of plots of activation rates for ion channels from NEURON mod file (example):
pynml-modchananalysis NaConductance -modFile NaConductance.mod
See here for more.
6) Home for existing functionality distributed in various places
Incorporate ChannelML2NeuroML2beta.xsl for updating ChannelML (coming soon...)
7) NEURON to NeuroML2
Scripts for converting NEURON to NeuroML2
-
Export morphologies (plus channels, soon). See here.
-
mod files - make best guess at initial NeuroML2 form (todo)
8) Export of images/movies from cell/networks
Files can be generated for POV-Ray which can be used to generate high resolution images and even sequences of images for creating movies. try:
pynml-povray -h
9) Tuning cell models in NeuroML 2
Builds on Neurotune and pyelectro. See here for example.
9) Planned functionality
Built in viewer of cells in 3D? Mayavi? More closely tied to PyNN?