What cell types are we usually following? What is the orientation of the cross-sectional views to the intact retina? same ? for 3D
Thanks, josephr
We have been working for 3 cells, and all of them are orientation selective ganglion cell. Refer this wiki page (still under construction; contribution welcome) for the information.
Originally, the EM images were obtained from the cross sections which are parallel to the optical axis (i.e., the retina tissue is sliced from the side). But once we get the stack of such images, cross sections from three orthogonal directions are all available and you are assigned with one of the three directions randomly chosen by the system.
"But once we get the stack of such images, cross sections from three orthogonal directions are all available and you are assigned with one of the three directions randomly chosen by the system."
I’d be very interested in knowing how that works exactly. 
I always figured once you sliced up a piece, there is no way of slicing it up in another direction afterwards.
I always figured once you sliced up a piece, there is no way of slicing it up in another direction afterwards.If you are assigned with one of the three directions randomly chosen, does that mean other people can get the same task but with a different orientation?
It would be very valuable to be able to switch orientations. It’s a lot easier when slices are perpendicular to the neuron you are tracing.
@whathecode,
There's no magic here. Just remember that we are playing with a stack of digital images, consisting of pixels. Schematically, it would look like the cube below. We can extract a "plane" of blocks from any of the 3 directions. This is only a matter of computation, and we don't have to glue the tissue and slice it again. :)
If you are assigned with one of the three directions randomly chosen, does that mean other people can get the same task but with a different orientation?It would be very valuable to be able to switch orientations. It's a lot easier when slices are perpendicular to the neuron you are tracing.
You got things correctly about these. It may be better if we could provide 3 orthogonal views to one user, but there are a few technical challenges. Please refer a previous discussion on the matter: