Topic: MoSeq

Quantitative methods that cluster videos of animal behavior into repeated syllables (sometimes called “motifs") have become fundamental tools for systems neuroscientists and neuroethologists. A popular approach is to use autoregressive hidden Markov models (ARHMMs) to identify behavioral syllables in the absence of labeled training data. This model, called MoSeq, has been shown to successfully segment depth-camera videos of mouse behavior into syllables that are familiar to a human observer. However, one issue with MoSeq is that it sorts behaviors which appear very similar to the human eye (e.

Behavior analysis typically relies on trial-based repetitions of constrained tasks. We aimed to uncover how age, pain, and itch influence the patterns of behavioral sequences and their transitions in freely behaving animals using MoSeq, a machine learning network analysis tool developed by the Datta lab. Mice of different ages were treated with chloroquine (itch) or formalin (pain) and placed in an open field arena and recorded for 30 minutes via a high-speed color camera and infrared depth camera. A trained autoregressive hidden Markov model classified the mouse movements into sub-second syll