Absorptions to current
The [@outerSegment class](Outer segment) includes methods that convert cone absorptions (R*) to photocurrent (pA). The method main method is implemented in the @osBioPhys subclass of @outerSegment, and it is a model developed by the Rieke Lab for primate cones.
In addition to characterizing the dynamics of the response, the method models light adaptation present at the level of the photoreceptors. Hence, some references to this model in our documents also refer to it as a 'Cone adaptation' model.
Finally, this model is used by the @osLinear class, as well. That class is implemented for conditions in which the background is relatively steady and stimuli are modest perturbations around the background (many psychophysical and physiological experiments have this character). We derive the linear impulse response functions that convert absorptions to photocurrent on the steady background by a quick computation using this model. That is why the @osLinear model varies with the background level.
The Rieke lab model is defined by a set of differential equations that capture different stages in the photoreceptor signal cascade.
The detailed explanation for each equation and the default values for the parameters can be found in the main citation.
The implementation uses three main functions
osAdaptSteadyState - Estimate steady-state current from background absorption rate
osAdaptTemporal - Calculate time varying current from time varying absorption
osAddNoise - Add noise to the current
These functions are implemented as part of the @osBioPhys subclass of the outerSegment class.
- osAdaptSteadyState computes response for a steady inputs stimulus. This function returns all intermediate state (PDE, cGMP, etc.) values together with the adapted outer segment membrane currents.
- osAdaptTemporal computes the temporal response for a time varying absorption rate by iteratively updates the discretized differential equations (numerical solution).
- osAddNoise does what its name suggests.
The script v_cmCurrentImpulse and several v_os* scripts illustrate and test properties of the photocurrent calculation.
The model parameters for the periphery and fovea differ (see t_osLinearize). Responses in the periphery are a bit faster, according to the data and hence the model parameters.
The comparison between ISETBIO implementations and Rieke's original script can be found here.
The disagreement is because derived formula is a little different for current computation. We will resolve this issue by discussing with Fred and compare the results with some real data.