Stephen Larson
Fahim Imam
This ontology is used to add the Definition class of annotation as in NIF.
The source of the definition can be defined with class from OBO-annotation
Computational Neuroscience Ontology
An ontology to describe the field of Computational Neurosciences
This work is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-sa/3.0/ or send a letter to Creative Commons, 444 Castro Street, Suite 900, Mountain View, California, 94041, USA.
Yann Le Franc
24/01/2012
INCF MultiScale Modeling Task Force
version 0.2.3
time constant
conductance-based model
Give a better definition. The problem is how to represent integrate and fire models with multiple conductances such as in Hill, 2005?
instantaneous rise and monexponential decay
facilitation model
spike response model with cumulative dynamic threshold
developmental learning
synapse model component
pre-synaptic element
reference the model from Froemke and Dan, 2002 described in Morrisson, 2008
suppression model
adaptive exponential integrate-and-fire
Connor-Stevens model
uniform
3D coordinates
Shinomoto model
volume
cellular model component
representation the different elements used to build cellular model types
difference of two exponentials
resting membrane potential
membrane potential at rest
Big question: Should it be considered as a parameter rather than a component?
rate
voltage-dependent current
find better definition?
one variable model
representation of the models including only one variable to describe the nerve cells behavior
learning rule
integrate-and-fire
polar coordinates
Rolstein model
This part of the classification is based upon the proposed categories in Morrisson, 2008
synaptic plasticity rule
represents the amount of time it takes for an excitable membrane to be ready for a second stimulus once it returns to its resting state following excitation
Could be considered as parameter?
refractory period
This class would contains models that describes the link between potassium channels and NMDA activation as described in hippocampus for instance.
cellular plasticity
represent the models that describe variations of the cellular intrinsic excitability due to activity.
alpha function
two variable model
representation of the models that include two variables to describe the nerve cells behavior
represents the models of currents reproducing a particular cellular behavior such as rebound or adaptation.
abstract current
Fitzhugh-Nagumo model
neurotransmittor diffusion
represents the intrinsic functional quality of the models
Example: oscillatory, bursting, synchronization, ...
functional quality
HindMarch and Rose model
pre-synaptic indice
reward-based learning
reinforcement learning
Should be linked to the cellular models with an object property such as "isInstanceOf"
cellular elements
representation of the cellular element model used to build the network model
synaptic plasticity
depression model
representation of the models that include three variables to describe the nerve cells behavior
three variable models
post-synaptic element
Wang-Buzsaki model
post-synaptic variable
resonate and fire
synaptic element
Should be linked to the "synaptic models" with object property as for cellular element
representation of the synaptic elements used to build a network model
user-defined
voltage dependent receptor activation
chemical synapse model
name
http://www.ebi.ac.uk/sbo/main/SBO:0000259
voltage
http://www.ebi.ac.uk/sbo/main/SBO:0000254
resistance
unsupervised learning
point
represents the models without specified morphology
Izhikevich model
non-linear
anatomically defined
plasticity model
represent the different model types existing to describe plasticity mechanisms
spike timing dependent plasticity model
two dimensional layout
relates to http://neurolex.org/wiki/Category:Two_dimensional_region
release mechanism
notation
capacitance
electrical synapse model
relates to http://neurolex.org/wiki/Category:Zero_dimensional_region
zero dimensional layout
threshold-based
represents the models considering the action potential initiation as a result of a thresholding opearion, to be considered in contrast with continuous mechanisms.
excitatory action
This class can be further subdivided into binary neuron, McCullogh neurons and all variants of the artificial neuron models.
artificial neuron model
representation of the abstract models of biological neurons.
random
integrate fire and burst
rate-based plasticity model
http://neurolex.org/wiki/Category:Three_dimensional_region
three dimensional layout
morphological quality
Similar to sao1057800815
membrane recovery mechanism
As a reference of variable u in Izhikevich model.
represents a qualitative feature of the model that can be either independent of the model specification or a property resulting of the particular model specification
model quality
inhibitory action
current
represents the models considering the action potential initiation as a continuous process, to be considered in contrast with threshold-based mechanisms.
continuous
indices
synaptic weight
concentration
Hodgkin and Huxley model
generalized integrate-and-fire
named model
spatial coordinates
long term plasticity model
spiking network model
number of elements
shunting action
time
represent models that have detailed spiking mechanism similar to Hodgkin and Huxley
detailed model
diameter
linear
post-synaptic indice
2D coordinates
short term plasticity model
network model
representation of the different model types used to represent neural networks
This class could become a superclass called dynamical models or dynamical system models in contrast with biophysical and threshold based models.
It should contain models like Fitzhugh-Nagumo, Morris-Lecar models.
representation of the models derived from Hodgkin and Huxley
simplified model
obsolete
representation of the values describing a particular parameter: name, notation, numerical value and units
parameter value
Instances of this class will further specify the general parameters described in the class "model parameters".
Abbott model
model description
represents the differents types of formats that can exist for describing a model (scientific paper, database entry, XML format, ...)
pre-synaptic variable
small world
spiking current
related to HindMarch and Rose model
Questionable
calcium dynamic
represents models describing the dynamic of the intracellular calcium concentration over time.
plasticity mechanism
represents the models of nerve cell morphology
morphology
representation of the models which represent the spiking mechanism as the result of a threshold crossing. The earliest example is the Lapicque's model (1907).
threshold-based model
maximal conductance
http://www.ebi.ac.uk/sbo/main/SBO:0000347
duration
model of network plasticity based on developmental rules that will change the structure of the network: Van Ooyen, Willshaw
structural plasticity
adaptive integrate-and-fire
Markram and Tsodyks model
biological correlate
represents the differents biological concepts and entities described by models.
This class represent the main entry point for NeuroLex/ NIF STD biological, structural and anatomical concepts. Model should be linked to this class with a dedicated property hasBiologicalCorrelate/isBiologicalCorrelateOf.
representation of the different parameters used to constrains models. The parameter can be either constant or variable.
This class contains general parameters (voltage, current, ...) rather than more specific parameters such membrane voltage, resting membrane potential, ...
These specific instances are then specified with the use of the class "Parameter values".
model parameter
feedforward
synaptic conductance dynamic
bursting current
related to HindMarch and Rose model. see Wikipedia for definition.
This term is questionable
spiking mechanism
represents the mechanisms generating action potential in nerve cells in model
reconstructed
Could a quality of the model rather than a component. Reference to the definition of quality by BFO
represents models with morphology build upon the reconstruction of a real nerve cell morphology using imaging techniques such as camera lucida or two-photon imaging.
frequency
biophysical model
representation of the class of models with a detailed spiking mechanisms representing the interaction between the fast sodium and slow potassium channels. The earliest example of this class is the Hogdkin and Huxley model of spike generation in the loligo axon.
network layout
This class is based on the assumption that models can represented as particular assemblies of components that can be either specific to the different level of description (cellular, synaptic and network) or generic.
model component
representation of the components used to build the different model types
homeostatic plasticity
synaptic delay
adaptation current
stimulation current
represents models describing input currents provided to the model via an electrode (intracellular or extracellular).
represents the models with an abstract morphology with multiple compartments as the Ball-and-stick model
multiple compartment
anatomical grouping
recurrent
area
link to SBO, Kisao, Teddy
represents the differents mathematical concepts used to represent models.
mathematical concept
biophysical model
model type
This general class includes the most common types of models classified based on the level of description of the nervous system.
Representation of the different types of models.
rebound current
functional grouping
represents models with one compartment
single compartment
artificial neural network model
corresponds to Kohonen maps, Multiple Layer perceptron, ...
represent the abstract neural network models based on artificial neuron models
rate function
intracellular dynamic
represents models of chemical signaling occuring inside the nerve cell
length
unit
leaky integrate-and-fire
connectivity pattern
should be linked with connectivity rule
represents well-defined categories of connectivity patterns obtained with specific connectivity rules
one dimensional layout
related to definition in NIF: http://neurolex.org/wiki/Category:One_dimensional_region
biochemically activated current
spiking model
representation of the models describing the properties of nerve cells, designed to accurately describe and predict biological processes. This is in contrast with artificial neurons.
triplet-based rule
Morris-Lecar model
rate model
represents the location of the current source on a neuronal model (somatic, primary dendrite, ...)
current localization
spike shape kernel
randomness
representation of the different types of models used to represent synapses
synapse model
phenomenological model
numerical value
exponential integrate-and-fire
cumulative spike response model
represents the distribution of the cell within a particular spatial layout
cellular distribution
pair-based rule
represent the statistical point process models used to reproduce particular feature of real neuron spiking patterns.
point process model
synaptic action
non linear spike generation current
represents an additional non-linear current ψ which leads to
a divergence of the potential toward infinity in a finite
time.
conductance-based model
In this case, the postsynaptic membrane potential is varying over time. It allows to capture phenomenon such as shunting inhibition
represents the different types of currents that can influence the trajectory of the membrane voltage
current type
The distinction between the different subclasses is based on the different "spiking mechanisms".
representation of the different model type used to describe nerve cells
cellular model
responsiveness kernel
generic model component
theta neuron model
spatial embedding
Models can be defined without any mention of space.
The model can be enriched with particular spatial embedding.
uniform
delay
http://www.ebi.ac.uk/sbo/main/SBO:0000225
spike response model SRM0
supervised learning
voltage and ionic dependent current
spike time
Hopfield network
connection element
network model component
represents the models describing the flow of current through chemically-gated channels or gap junctions (connexons).
synaptic current
dynamic threshold
This particularity is created by keeping the postsynaptic membrane voltage constant in the current equation => isyn=gsyn*(Cte-Erev)
current-based model
quadratic integrate-and-fire
connectivity rule
cellular grouping
mixed grouping
abstract
ionic dependent current
spike response model
balanced random network
probability
biochemical dynamic
represents models describing the dynamic of the intracellular biochemical reactions and molecular participant concentrations over time.
synaptic weight
ionic current
represents the models describing the flow of current through uniquely voltage-gated channels
spike response model
representation of the models that correspond to a generalization of the leaky integrate-and-fire model and give a simple description of action potential generation in neurons
represents the spatial distribution of current sources (ionic channels, synaptic channels, stimulation electrode) on a compartment
current distribution
fixed threshold
isVariableOf
hasBiologicalCorrelate
isBiologicalCorrelateOf
hasComponent
isParameterOf
isComponentOf
hasParameter
isParameterValueOf
hasParameterValue
hasVariable