TY - JOUR
T1 - SteftR
T2 - A hybrid versatile method for state estimation and feature extraction from the trajectory of animal behavior
AU - Yamazaki, Shuhei J.
AU - Ohara, Kazuya
AU - Ito, Kentaro
AU - Kokubun, Nobuo
AU - Kitanishi, Takuma
AU - Takaichi, Daisuke
AU - Yamada, Yasufumi
AU - Ikejiri, Yosuke
AU - Hiramatsu, Fumie
AU - Fujita, Kosuke
AU - Tanimoto, Yuki
AU - Yamazoe-Umemoto, Akiko
AU - Hashimoto, Koichi
AU - Sato, Katsufumi
AU - Yoda, Ken
AU - Takahashi, Akinori
AU - Ishikawa, Yuki
AU - Kamikouchi, Azusa
AU - Hiryu, Shizuko
AU - Maekawa, Takuya
AU - Kimura, Koutarou D.
N1 - Funding Information:
We thank Drs. André Brown, Katsuyoshi Matsushita, Ken-ichi Hironaka, Takuma Degawa, Daisuke Yamamoto, Soh Kohatsu, and KDK laboratory members for suggestions and comments on this work. Nematode strains were provided by the Caenorhabditis Genetics Center (funded by the NIH Office of Research Infrastructure Programs P40 OD010440) and by the National BioResource Project funded by the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan.
Funding Information:
This work was supported by the Interdisciplinary graduate school program for systematic understanding of health and disease (for SJY), by the Tohoku Ecosystem-Associated Marine Science (for KS), by 58th Japanese Antarctic Research Expedition (for KI, NK, and AT) by KAKENHI JP16K16185 (for NK), JP25249020 and JP16H06536 (for KH), JP24681006, JP16H01769, JP16H06541 (for KY), 17H05983 (for AT), JP16H04655, JP18H05069, and JP17K19450 (for AK), JP16H06542 (for SH), JP16H06539 (for TM), JP16H06545 (for KDK) from the MEXT, by PRESTO 11008 (for SH) by JST, and by the Osaka University Co-Creation Program (for TM and KDK).
Publisher Copyright:
Copyright © 2019 Yamazaki, Ohara, Ito, Kokubun, Kitanishi, Takaichi, Yamada, Ikejiri, Hiramatsu, Fujita, Tanimoto, Yamazoe-Umemoto, Hashimoto, Sato, Yoda, Takahashi, Ishikawa, Kamikouchi, Hiryu, Maekawa and Kimura. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
PY - 2019
Y1 - 2019
N2 - Animal behavior is the final and integrated output of brain activity. Thus, recording and analyzing behavior is critical to understand the underlying brain function. While recording animal behavior has become easier than ever with the development of compact and inexpensive devices, detailed behavioral data analysis requires sufficient prior knowledge and/or high content data such as video images of animal postures, which makes it difficult for most of the animal behavioral data to be efficiently analyzed. Here, we report a versatile method using a hybrid supervised/unsupervised machine learning approach for behavioral state estimation and feature extraction (STEFTR) only from low-content animal trajectory data. To demonstrate the effectiveness of the proposed method, we analyzed trajectory data of worms, fruit flies, rats, and bats in the laboratories, and penguins and flying seabirds in the wild, which were recorded with various methods and span a wide range of spatiotemporal scales—from mm to 1,000 km in space and from sub-seconds to days in time. We successfully estimated several states during behavior and comprehensively extracted characteristic features from a behavioral state and/or a specific experimental condition. Physiological and genetic experiments in worms revealed that the extracted behavioral features reflected specific neural or gene activities. Thus, our method provides a versatile and unbiased way to extract behavioral features from simple trajectory data to understand brain function.
AB - Animal behavior is the final and integrated output of brain activity. Thus, recording and analyzing behavior is critical to understand the underlying brain function. While recording animal behavior has become easier than ever with the development of compact and inexpensive devices, detailed behavioral data analysis requires sufficient prior knowledge and/or high content data such as video images of animal postures, which makes it difficult for most of the animal behavioral data to be efficiently analyzed. Here, we report a versatile method using a hybrid supervised/unsupervised machine learning approach for behavioral state estimation and feature extraction (STEFTR) only from low-content animal trajectory data. To demonstrate the effectiveness of the proposed method, we analyzed trajectory data of worms, fruit flies, rats, and bats in the laboratories, and penguins and flying seabirds in the wild, which were recorded with various methods and span a wide range of spatiotemporal scales—from mm to 1,000 km in space and from sub-seconds to days in time. We successfully estimated several states during behavior and comprehensively extracted characteristic features from a behavioral state and/or a specific experimental condition. Physiological and genetic experiments in worms revealed that the extracted behavioral features reflected specific neural or gene activities. Thus, our method provides a versatile and unbiased way to extract behavioral features from simple trajectory data to understand brain function.
KW - Behavioral states
KW - Calcium imaging
KW - Feature extraction
KW - Genetic screening
KW - Navigation
KW - Quantitative behavioral analysis
UR - http://www.scopus.com/inward/record.url?scp=85068484281&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85068484281&partnerID=8YFLogxK
U2 - 10.3389/fnins.2019.00626
DO - 10.3389/fnins.2019.00626
M3 - Article
AN - SCOPUS:85068484281
VL - 13
JO - Frontiers in Neuroscience
JF - Frontiers in Neuroscience
SN - 1662-4548
IS - JUN
M1 - 626
ER -