Identifying behavioral structure from deep variational embeddings of animal motion.

Identifying behavioral structure from deep variational embeddings of animal motion.
复制标题

DOI:
10.1038/s42003-022-04080-7
复制
发表时间:
2022-11-18
影响因子:
5.9
通讯作者:
Bauer, Pavol
Bauer, Pavol
中科院分区:
生物学2区
文献类型:
--
作者:
Luxem, Kevin;Mocellin, Petra;Fuhrmann, Falko;Kuersch, Johannes;Miller, Stephanie R.;Palop, Jorge J.;Remy, Stefan;Bauer, Pavol

文献摘要

参考文献

被引文献

相似文献

行为分层组织的量化和检测是神经科学的主要挑战。无标记姿态估计的最新进展使得动物运动的高维时空行为动力学可视化成为可能。然而,需要强大而可靠的技术方法来揭示这些数据中的底层结构,并将行为分割成离散的分层组织的主题。在这里,我们提出了一个无监督的概率深度学习框架,该框架从动物运动的深度变分嵌入(VAME)中识别行为结构。通过使用 β 淀粉样变性小鼠模型作为用例,我们表明 VAME 不仅可以识别离散的行为主题,还可以捕获该主题使用的分层表示。该方法允许将基序分组为群落,并检测单个小鼠群体的群落特异性基序使用的差异,这是人类视觉观察无法检测到的。因此,我们提出了一种稳健的动物运动分割方法,适用于各种实验设置、模型和条件,无需监督或先验的人为干扰。动物运动的变异嵌入能够自动发现离散的行为主题及其潜在的动态。
Quantification and detection of the hierarchical organization of behavior is a major challenge in neuroscience. Recent advances in markerless pose estimation enable the visualization of high-dimensional spatiotemporal behavioral dynamics of animal motion. However, robust and reliable technical approaches are needed to uncover underlying structure in these data and to segment behavior into discrete hierarchically organized motifs. Here, we present an unsupervised probabilistic deep learning framework that identifies behavioral structure from deep variational embeddings of animal motion (VAME). By using a mouse model of beta amyloidosis as a use case, we show that VAME not only identifies discrete behavioral motifs, but also captures a hierarchical representation of the motif’s usage. The approach allows for the grouping of motifs into communities and the detection of differences in community-specific motif usage of individual mouse cohorts that were undetectable by human visual observation. Thus, we present a robust approach for the segmentation of animal motion that is applicable to a wide range of experimental setups, models and conditions without requiring supervised or a-priori human interference. Variational embeddings of animal motion enable automated discovery of discrete behavioural motifs as well as their latent dynamics.
DOI: 10.1038/s41592-021-01106-6
发表时间: 2021-05
期刊: Nature methods
影响因子: 48
作者:
Dunn TW;Marshall JD;Severson KS;Aldarondo DE;Hildebrand DGC;Chettih SN;Wang WL;Gellis AJ;Carlson DE;Aronov D;Freiwald WA;Wang F;Ölveczky BP
通讯作者: Ölveczky BP
DOI: 10.1038/s41593-019-0460-x
发表时间: 2019-09-01
影响因子: 25
作者:
Chaudhuri, Rishidev;Gercek, Berk;Fiete, Ila
通讯作者: Fiete, Ila
DOI: 10.1186/s12915-018-0494-7
发表时间: 2018-02-23
期刊: BMC biology
影响因子: 5.4
作者:
Berman GJ
通讯作者: Berman GJ
DOI: 10.7554/elife.63377
发表时间: 2021-09-02
期刊: eLife
影响因子: 7.7
作者:
Bohnslav JP;Wimalasena NK;Clausing KJ;Dai YY;Yarmolinsky DA;Cruz T;Kashlan AD;Chiappe ME;Orefice LL;Woolf CJ;Harvey CD
通讯作者: Harvey CD
DOI: 10.1007/s11357-016-9929-7
发表时间: 2016-08-01
期刊: AGE
影响因子: --
作者:
Huang, Huang;Nie, Sipei;Xiao, Ming
通讯作者: Xiao, Ming