About - DeNardoLab/BehaviorDEPOT GitHub Wiki

BehaviorDEPOT

An open-source behavioral analysis toolbox for rodents designed to first compile and clean point-tracking output from DeepLabCut (or similar software), and then classify behavioral epochs using custom behavior classifiers. BehaviorDEPOT support use of custom ROIs and time-locked cues and provides analysis of the intersection between tracked behavior(s) and the applied spatiotemporal filters.

BehaviorDEPOT is designed to be customizable and easily tunable to each individual behavior task or environment. We encourage users (and provide guidance) to adapt and customize the code, including building classifiers for new behaviors and/or tasks.

Motivation

We wanted a tool that seamlessly takes in output from DeepLabCut, smoothes out hiccups in tracking, and identifies bouts of specific behaviors, in particular freezing, as well as helpful metrics about animal movement.

Credits

BehaviorDEPOT was developed by Chris Gabriel, Benita Jin, Zachary Zeidler, and Laura DeNardo. If you find this tool useful, please cite the work: {will be updated upon publication}