Meal pattern analysis

[Ames Sutton]

Hi all,

I have an analysis question from FED data that I've been pondering and I'm wondering if others have run into something similar/have any suggestions. I have a paradigm with three groups in which I'm trying to determine if their meal patterns are different. In the FED ELife paper from Lex's group, it looks like meal pattern analysis was done by looking at the histogram of pellets and deciding a time point for meals to be an interpellet interval time under that at which the majority of pellets were eaten (see screenshot). 

My mice are food restricted, which shifts the curve a bit. I've been using an arbitrary value of retrieval time for the pellet (t=45 seconds, see attached .jpeg) illustrated here with a vertical line in the histogram. Does anyone have a suggestion for an alternative approach? Would it be more suitable to use the peak of the histogram (e.g. median/mean) for meal characterization?

Thanks in advance!

Ames

--

Ames Sutton, PhD

Postdoctoral Fellow, Krashes Lab

NIDDK, National Institutes of Health

cell: 317-435-4631

[Matias Andina]

Hello,

I am also interested in this, and my mice are also food deprived. I have toyed around with "burst" detection (basically using the log interpellet interval to detect bursts of "pellet eating activity"). I haven't done a full analysis but I am happy to share code and discuss whether people see a lot of variability in metrics such as the number of bouts, the length of the bouts. I think these metrics can be fickle and might not reflect what we want them to reflect, but might still be a good idea to try!

This was a quick and dirty and plot I made back when I was checking this, with pellet events as ticks and the red lines connecting the detected bouts. Again, happy to share code if somebody interested!

--
You received this message because you are subscribed to the Google Groups "FEDforum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fedforum+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fedforum/CAM0mzO826f_wGvytLovTBWks-BXUCzLKOP8dx2%2BUocRzgkeRTg%40mail.gmail.com.

[Ames Sutton]

I'd love to try out that code! Thanks, Matias. I think the log IPI could be an interesting approach. I'd be interested to hear others takes as well.

Thanks again,

Ames