Notes, assignments, and code for NEUROBIO 735 (Spring 2018).
1/10 – 2/8:
Wednesday, Thursday
3:00 – 4:30
DIBS conference room
In this session, we’ll continue our analysis of orientation and motion tuning in the calcium imaging sample from mouse V1. Our first step will be to refactor our current tuning analysis into a function, making it easier to swap out our method for determining preferred orientation while keeping the rest of our code the same (aka, the strategy pattern). The second step will be to write code that calculates tuning across the entire image.
Before we start, let’s clean up our code:
n
values, use, e.g., parula(n)
(parula is the name of the default colormap).colormap(cmap)
, where cmap
is a colormap.caxis
function. I recommend simply creating a colormap with as many colors as you need and plotting the image by assigning each pixel an integer from 1 to the number of stimuli. This should work as expected.For the rest of this exercise, we’ll be improving our methods of calculating and plotting the pixel tuning.
If we want to consider our pixel as tuned, or a particular value the preferred value, it will help to have a sense of how confident we can be in our tuning curve.
It’s clear from both biology and the data at hand that not every pixel is even responsive to the stimuli. We would like a way to “opt out” of coloring a pixel with a preferred orientation in cases where that doesn’t make sense.