Data from: Inferring neural circuit properties from optogenetic stimulation

Optogenetics has become an important tool for perturbing neural circuitry
with unparalleled temporal precision and cell-type specificity. However,
direct activation of a specific subpopulation of neurons can rapidly
modulate the activity of other neurons within the network and may lead to
unexpected and complex downstream effects. Here, we have developed a
biologically-constrained computational model that exploits these
non-intuitive network responses in order to gain insight into underlying
properties of the network. We apply this model to data recorded during
optogenetic stimulation in the primary visual cortex of the alert macaque.
In these experiments, we found that optogenetic depolarization of
excitatory neurons often suppressed neuronal responses, consistent with
engagement of normalization circuitry. Our model suggests that the
suppression seen in these responses may be mediated by slow excitatory and
inhibitory conductance channels. Furthermore, the model predicted that the
response of the network to optogenetic perturbation depends critically on
the relationship between inherent temporal properties of the network and
the temporal properties of the opsin. Consistent with model predictions,
stimulation of the C1V1TT opsin, an opsin with a fast time constant
(tau=45 ms), caused faster and stronger suppressive effects after laser
offset, as compared to stimulation of the slower C1V1T opsin (tau=60ms).
This work illustrates how the non-intuitive network responses that result
from optogenetic stimulation can be exploited to gain insight regarding
network properties that underlie fundamental neuronal computations, such
as normalization. This novel hybrid opto-theoretical approach can thus
enhance the power of optogenetics to dissect complex neural circuits.