Overview

Powerful Features.
Gain insights into your
experimental data

NetPyNE is an open-source Python package to facilitate the development, parallel simulation, analysis, and optimization of brain circuit models using the NEURON simulator.
Easy specification
Define models through an intuitive, standardized, declarative Python-based format, officially endorsed by INCF as a standard
Simulation-ready models
Converts high-level specification into NEURON network instance ready for parallel simulation.
Built-in analysis & visualization
Wide range of functions to explore connectivity, voltages/currents, spikes, LFPs, EEG, stats, information measures, ...

Define. Set model properties accross all scales.

Declarative language
Standardized, high-level specification of biological parameters (programmatic or via Graphical Web App)
Multiscale
Include multiple scales from intra- and extra-cellular molecular reaction-diffusion (RxD) to cellular to circuit level.
Cells and populations
Define cell morphologies, biophysics, density, spatial distribution, ...
Connectivity & stimulation
Specify rule-based, complex connectivity and stimulation patterns at the population, cell, and dendritic level
Importing
Use existing cell models defined in HOC, Python or NeuroML

Simulate.
Run parallel simulations.

Efficient implementation
Automated, robust and well-tested backend NEURON implementation prevents coding errors and inefficiencies.
Parallel simulation
Easily set up and run parallel simulations, including distributing cells and gathering data from computing nodes.
Network instance
Generate Python hierarchical structure with NEURON network objects directly from declarative specification.

Analyse. Visualize & explore your network and simulation output.

Connectivity
Connectivity matrix or bar plots showing probability, weights or number of connections
Spiking activity & statistics
Raster plot, histograms, spectrograms, average rate, variability, & synchrony measures.
Neurons 3D explorer
State-of-the-art neuron morphology 3D viewer and 3D shape plot with color-coded variables (e.g number of synapses)
Intrinsic cell properties
Analysis and plotting of membrane voltages, ionic/synaptic currents, conductances, molecular concentrations, etc
Network 2D representation
2D representation of network cell locations and connections
Information transfer
Normalized transfer entropy and spectral Granger Causality
LFP and CSD
Local field potentials (LFP) plots (time-resolved, power spectra and spectrogram), and current source density (CSD) analysis
EEG
Scalp electroencephalogram (EEG) signals based on neuron current dipole moments and head volume conduction model (using LFPykit)
blur_on
Parameter exploration/optimization
check
Automatically explore or optimize model parameters via grid search or evolutionary algorithms
check
Easy specification of parameters and range of values to explore
check
Pre-defined, configurable setups to automatically submit jobs in multicore machines (bulletin board) or supercomputers (SLURM or PBS Torque)
share
Data Sharing
check
Save/load high-level specs, network instance, simulation configuration, and simulation results
check
Multiple formats supported: JSON, Pickle, Matlab, HDF5
check
Export/import to/from NeuroML and SONATA standardized formats for neural models

Explore many models developed by our growing community.

Explore the growing list of NetPyNE models being developed by the user community, and use them as a starting point, to get ideas or to borrow components. These include models of many different brain regions (e.g. cortex, thalamus, hippocampus, cerebellum), disease (e.g. epilepsy, schizophrenia, ischemia), treatments (e.g. ketamine, TMS/tDCS stimulation) and other brain phenomena (e.g. oscillatory coding, sensorimotor learning, memory retrieval).

Open Models Library
Multiscale model of primary motor cortex (M1) circuits
Auditory thalamocortical detailed model
Somatosensory thalamocortical detailed model
Motor, somatosensory and auditory thalamocortical circuit models
CA3 network
Potjans & Diesmann cortical microcircuit model
M1 model based on Izhikevich neurons
Thalamocortical network with epilepsy
Alterations of schizophrenia-associated genes on gamma band oscillations
Single cell (M1 PT) batch parameter exploration example
Pyramidal-tract project (PT) corticospinal cell in M1
Spiking neuronal network model of visual-motor cortex playing a virtual racket-ball game
Spiking neuronal networks performing motor control
A simple sensorimotor network with STDP and RL that learns to drive a simple virtual arm
CA1 microcircuit exhibiting phase-amplitude coupling
Using NeuroML import of a simple network
Computer model of the spinal dorsal horn
Neurorobotics model of Parkinson's Disease
Computational model of the basal ganglia-thalamus-cortex
Model of Parkinson’s Disease tuned to reproduce oscillatory behavior
More models

Enjoy interfaces and integration with many tools, platforms and standards.

NetPyNE's growing ecosystem includes interfaces with many tools, e.g., coreNEURON, LFPy, The Virtual Brain and SciUnit; standards in the field, e.g., NeuroML and SONATA; and integration with multiple platforms, e.g., Open Source Brain (OSB), EBRAINS and the Neuroscience Gateway (NSG).

Loved by the neuroscientists around the world

Andreas Neef
Göttingen Center for Neurosciences
It works very well, scales nicely (we tested between 8 and 192 cores) and was in general easy to implement. Thanks for the great tool. I intend to use it in the near future.
Martha Bagnall
Washington University School of Medicine
I'm really excited about your software! I've coded in NEURON and it's a real relief to use a higher-level language.
Mohamed Sherif
Brown University
I have an undergraduate who really likes NetPyne, she said "your code turns into neurons" which I think is outstanding - she doesn't have any computational neuroscience experience before.
Aman Aberra
Duke University
I recently moved the network to NetPyNe to more easily parallelize on our cluster and try out different single-cell models/mechanisms, which has been great.
Parvin Zarei
University of Melbourne
Thank you so much for developing NetPyNE. It made implementing so many things much easier. Such an awesome tool.
Logan Bayer
Sent in the forum
NetPyNE has been incredible to learn and work in!
Andreas Neef
Göttingen Center for Neurosciences
It works very well, scales nicely (we tested between 8 and 192 cores) and was in general easy to implement. Thanks for the great tool. I intend to use it in the near future.
Martha Bagnall
Washington University School of Medicine
I'm really excited about your software! I've coded in NEURON and it's a real relief to use a higher-level language.
Mohamed Sherif
Brown University
I have an undergraduate who really likes NetPyne, she said "your code turns into neurons" which I think is outstanding - she doesn't have any computational neuroscience experience before.
Aman Aberra
Duke University
I recently moved the network to NetPyNe to more easily parallelize on our cluster and try out different single-cell models/mechanisms, which has been great.
Parvin Zarei
University of Melbourne
Thank you so much for developing NetPyNE. It made implementing so many things much easier. Such an awesome tool.
Logan Bayer
Sent in the forum
NetPyNE has been incredible to learn and work in!