bc / retina

Tools for visualizing and comparing data from vertebrate retinas
14 stars 3 forks source link

Quantifying retinal topography #19

Closed lschmitz closed 6 years ago

lschmitz commented 7 years ago

Users will benefit from the ability to quantify retinal landmarks. For an example, see http://jov.arvojournals.org/article.aspx?articleid=2191987.

Let's start simple. Where is the location of the highest density in latitude and longitude relative to the pole?

At least two approaches are feasible. Calculate peak position on the basis of the real counts (1), or estimated from the fitted surface (2).

Approach 1 may work like this:

findPeak <- function(my_retina){ coord <- my_retina$trimmed_data peak_temp <- coord[which.max(coord$z),] peak <- data.frame(latitude=abs(peak_temp[,1]), longitude=peak_temp[,2], peak=peak_temp[,3]) return(peak) }

Approach 2 should work with predictSurface.

briancohn commented 7 years ago

how do people most often refer to the peak location?

briancohn commented 7 years ago

and how is it used with statistics? Do people say that the peak longitude is significantly shifted in species A vs Species B?

lschmitz commented 7 years ago

Peak RGC density, usually associated with area centralis or fovea. Its position may be related to habitat/ecology, so one would test across species or across an ontogenetic series.