SUPR
Unsupervised classification of behavior from pose estimation data
Dnr:

NAISS 2025/22-711

Type:

NAISS Small Compute

Principal Investigator:

Jakub Mlost

Affiliation:

Stockholms universitet

Start Date:

2025-05-15

End Date:

2026-06-01

Primary Classification:

10203: Bioinformatics (Computational Biology) (Applications at 10610)

Webpage:

Allocation

Abstract

Understanding animal behavior is crucial for decoding how the brain works, modeling neurological disorders, and assessing treatments. Recent advances in pose-estimation tools like DeepLabCut and SLEAP have revolutionized behavioral analysis by enabling precise tracking of animal body movements. However, these tools do not automate behavioral classification. By bridging the gap between movement tracking and behavior classification, unsupervised learning algorithms promise to revolutionize neuroscience by providing scalable, unbiased methods for behavioral analysis and enabling researchers to uncover previously unknown patterns in animal behavior. This study implements one of the most recent unsupervised learning pipelines, Keypoint-MoSeq, to create an ethogram of mouse behavior following pharmacological manipulations of the serotonin system. This pipeline employs an autoregressive hidden Markov model to detect recurring motifs of an animal's posture evolution over time. Results of this project will bring a better understanding of how serotonergic drugs, such as antidepressants or psychedelics, affect behavior.