Keypoint-MoSeq: parsing behavior by linking point tracking to pose dynamics.

Autor:	Weinreb C; Department of Neurobiology, Harvard Medical School, Boston, MA, USA., Pearl J; Department of Neurobiology, Harvard Medical School, Boston, MA, USA., Lin S; Department of Neurobiology, Harvard Medical School, Boston, MA, USA., Osman MAM; Department of Neurobiology, Harvard Medical School, Boston, MA, USA., Zhang L; Department of Electrical Engineering, Stanford University, Stanford, CA, USA.; Wu Tsai Neurosciences Institute, Stanford University, Stanford, CA, USA., Annapragada S; Department of Neurobiology, Harvard Medical School, Boston, MA, USA., Conlin E; Department of Neurobiology, Harvard Medical School, Boston, MA, USA., Hoffman R; Department of Neurobiology, Harvard Medical School, Boston, MA, USA., Makowska S; Department of Neurobiology, Harvard Medical School, Boston, MA, USA., Gillis WF; Department of Neurobiology, Harvard Medical School, Boston, MA, USA., Jay M; Department of Neurobiology, Harvard Medical School, Boston, MA, USA., Ye S; Brain Mind and Neuro-X Institute, School of Life Sciences, Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland., Mathis A; Brain Mind and Neuro-X Institute, School of Life Sciences, Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland., Mathis MW; Brain Mind and Neuro-X Institute, School of Life Sciences, Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland., Pereira T; Salk Institute for Biological Studies, La Jolla, USA., Linderman SW; Wu Tsai Neurosciences Institute, Stanford University, Stanford, CA, USA.; Department of Statistics, Stanford University, Stanford, CA, USA., Datta SR; Department of Neurobiology, Harvard Medical School, Boston, MA, USA.
Jazyk:	angličtina
Zdroj:	BioRxiv : the preprint server for biology [bioRxiv] 2023 Dec 23. Date of Electronic Publication: 2023 Dec 23.
DOI:	10.1101/2023.03.16.532307
Abstrakt:	Keypoint tracking algorithms have revolutionized the analysis of animal behavior, enabling investigators to flexibly quantify behavioral dynamics from conventional video recordings obtained in a wide variety of settings. However, it remains unclear how to parse continuous keypoint data into the modules out of which behavior is organized. This challenge is particularly acute because keypoint data is susceptible to high frequency jitter that clustering algorithms can mistake for transitions between behavioral modules. Here we present keypoint-MoSeq, a machine learning-based platform for identifying behavioral modules ("syllables") from keypoint data without human supervision. Keypoint-MoSeq uses a generative model to distinguish keypoint noise from behavior, enabling it to effectively identify syllables whose boundaries correspond to natural sub-second discontinuities inherent to mouse behavior. Keypoint-MoSeq outperforms commonly used alternative clustering methods at identifying these transitions, at capturing correlations between neural activity and behavior, and at classifying either solitary or social behaviors in accordance with human annotations. Keypoint-MoSeq therefore renders behavioral syllables and grammar accessible to the many researchers who use standard video to capture animal behavior. Competing Interests: Competing interests S.R.D. sits on the scientific advisory boards of Neumora and Gilgamesh Therapeutics, which have licensed or sub-licensed the MoSeq technology.
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu