B-SOiD, an open-source unsupervised algorithm for identification and fast prediction of behaviors

• Published in: Nature communications Vol. 12; no. 1; pp. 5188 - 13
• Main Authors: Hsu, Alexander I., Yttri, Eric A.
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 31.08.2021; Nature Publishing Group; Nature Portfolio
• Subjects: 631/1647/2198; 631/378/1689; 631/378/2632; Algorithms; Animal behavior; Animal models; Animals; Behavior; Behavior, Animal; Behavioral Sciences - instrumentation; Behavioral Sciences - methods; Biomechanical Phenomena; Female; Human bias; Humanities and Social Sciences; Humans; Kinematics; Learning algorithms; Localization; Machine Learning; Male; Mice - physiology; Mice, Inbred C57BL; Movement disorders; multidisciplinary; Science; Science (multidisciplinary); Segmentation; Software; Temporal resolution; Tracking
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-021-25420-x

Studying naturalistic animal behavior remains a difficult objective. Recent machine learning advances have enabled limb localization; however, extracting behaviors requires ascertaining the spatiotemporal patterns of these positions. To provide a link from poses to actions and their kinematics, we developed B-SOiD - an open-source, unsupervised algorithm that identifies behavior without user bias. By training a machine classifier on pose pattern statistics clustered using new methods, our approach achieves greatly improved processing speed and the ability to generalize across subjects or labs. Using a frameshift alignment paradigm, B-SOiD overcomes previous temporal resolution barriers. Using only a single, off-the-shelf camera, B-SOiD provides categories of sub-action for trained behaviors and kinematic measures of individual limb trajectories in any animal model. These behavioral and kinematic measures are difficult but critical to obtain, particularly in the study of rodent and other models of pain, OCD, and movement disorders. The study of naturalistic behaviour using video tracking is challenging. Here the authors develop a system, B-SOiD which allows automated behavioural tracking and segmentation of video of movements tested in mice, flies and humans.