Compartment#

class dendrify.compartment.Compartment(name, model='passive', **kwargs)[source]#

Bases: object

A class that automatically generates and handles all differential equations and parameters needed to describe a single compartment and any currents (synaptic, dendritic, noise) passing through it.

Parameters:

  • name (str) – A unique name used to tag compartment-specific equations and parameters. It is also used to distinguish the various compartments belonging to the same NeuronModel.

  • model (str, optional) – A keyword for accessing Dendrify’s library models. Custom models can also be provided but they should be in the same formattable structure as the library models. Available options: 'passive' (default), 'adaptiveIF', 'leakyIF', 'adex'.

  • kwargs (Quantity, optional) – Kwargs are used to specify important electrophysiological properties, such as the specific capacitance or resistance. For more information see: EphysProperties.

Examples

>>> # specifying equations only:
>>> compX = Compartment('nameX', 'leakyIF')
>>> # specifying equations and ephys properties:
>>> compY = Compartment('nameY', 'adaptiveIF', length=100*um, diameter=1*um,
>>>                     cm=1*uF/(cm**2), gl=50*uS/(cm**2))

Attributes:

area

A compartment's surface area (open cylinder) based on its length and diameter.

capacitance

A compartment's absolute capacitance based on its specific capacitance (cm) and surface area.

equations

All differential equations that have been generated for a single compartment.

g_leakage

A compartment's absolute leakage conductance based on its specific leakage conductance (gl) and surface area.

parameters

All parameters that have been generated for a single compartment.

Methods:

connect

Allows the connection (electrical coupling) of two compartments.

noise

Adds a stochastic noise current.

synapse

Adds synaptic currents equations and parameters.
property area#

A compartment’s surface area (open cylinder) based on its length and diameter.

Return type:

Quantity

property capacitance#

A compartment’s absolute capacitance based on its specific capacitance (cm) and surface area.

Return type:

Quantity

connect(other, g='half_cylinders')[source]#

Allows the connection (electrical coupling) of two compartments.

Parameters:

  • other (Compartment) – Another compartment.

  • g (str or Quantity, optional) – The coupling conductance. It can be set explicitly or calculated automatically (provided all necessary parameters exist). Available options: 'half_cylinders' (default), 'cylinder_<compartment name>'.

Warning

The automatic approaches require that both compartments to be connected have specified length, diameter and axial resistance.

Examples

>>> compX, compY = Compartment('x', **kwargs), Compartment('y', **kwargs)
>>> # explicit approach:
>>> compX.connect(compY, g=10*nS)
>>> # half cylinders (default):
>>> compX.connect(compY)
>>> # cylinder of one compartment:
>>> compX.connect(compY, g='cylinder_x')
property equations#

All differential equations that have been generated for a single compartment.

Return type:

str

property g_leakage#

A compartment’s absolute leakage conductance based on its specific leakage conductance (gl) and surface area.

Return type:

Quantity

noise(tau=20. * msecond, sigma=3. * pamp, mean=0. * amp)[source]#

Adds a stochastic noise current. For more information see the Noise section: of Models and neuron groups

Parameters:

  • tau (Quantity, optional) – Time constant of the Gaussian noise, by default 20*ms

  • sigma (Quantity, optional) – Standard deviation of the Gaussian noise, by default 3*pA

  • mean (Quantity, optional) – Mean of the Gaussian noise, by default 0*pA

property parameters#

All parameters that have been generated for a single compartment.

Return type:

dict

synapse(channel=None, pre=None, g=None, t_rise=None, t_decay=None, scale_g=False)[source]#

Adds synaptic currents equations and parameters. When only the decay time constant t_decay is provided, the synaptic model assumes an instantaneous rise of the synaptic conductance followed by an exponential decay. When both the rise t_rise and decay t_decay constants are provided, synapses are modelled as a sum of two exponentials. For more information see: Modeling Synapses by Arnd Roth & Mark C. W. van Rossum

Parameters:

  • channel (str) – Synaptic channel type. Available options: 'AMPA', 'NMDA', 'GABA', by default None

  • pre (str) – A unique name to distinguish synapses of the same type coming from different input sources, by default None

  • g (Quantity) – Maximum synaptic conductance, by default None

  • t_rise (Quantity) – Rise time constant, by default None

  • t_decay (Quantity) – Decay time constant, by default None

  • scale_g (bool, optional) – Option to add a normalization factor to scale the maximum conductance at 1 when synapses are modelled as a difference of exponentials (have both rise and decay kinetics), by default False.

Examples

>>> comp = Compartment('comp')
>>> # adding an AMPA synapse with instant rise & exponential decay:
>>> comp.synapse('AMPA', g=1*nS, t_decay=5*ms, pre='X')
>>> # same channel, different conductance & source:
>>> comp.synapse('AMPA', g=2*nS, t_decay=5*ms, pre='Y')
>>> # different channel with both rise & decay kinetics:
>>> comp.synapse('NMDA', g=1*nS, t_rise=5*ms, t_decay=50*ms, pre='X')