Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Apr 15;13(4):bio060237.
doi: 10.1242/bio.060237. Epub 2024 Apr 23.

DeepLabCut-based daily behavioural and posture analysis in a cricket

Shota Hayakawa  1 , Kosuke Kataoka  2 , Masanobu Yamamoto  3 , Toru Asahi  1   2   4 , Takeshi Suzuki  3
Affiliations

DeepLabCut-based daily behavioural and posture analysis in a cricket

Shota Hayakawa et al. Biol Open. .

Abstract

Circadian rhythms are indispensable intrinsic programs that regulate the daily rhythmicity of physiological processes, such as feeding and sleep. The cricket has been employed as a model organism for understanding the neural mechanisms underlying circadian rhythms in insects. However, previous studies measuring rhythm-controlled behaviours only analysed locomotive activity using seesaw-type and infrared sensor-based actometers. Meanwhile, advances in deep learning techniques have made it possible to analyse animal behaviour and posture using software that is devoid of human bias and does not require physical tagging of individual animals. Here, we present a system that can simultaneously quantify multiple behaviours in individual crickets - such as locomotor activity, feeding, and sleep-like states - in the long-term, using DeepLabCut, a supervised machine learning-based software for body keypoints labelling. Our system successfully labelled the six body parts of a single cricket with a high level of confidence and produced reliable data showing the diurnal rhythms of multiple behaviours. Our system also enabled the estimation of sleep-like states by focusing on posture, instead of immobility time, which is a conventional parameter. We anticipate that this system will provide an opportunity for simultaneous and automatic prediction of cricket behaviour and posture, facilitating the study of circadian rhythms.

Keywords: Circadian rhythm; Cricket; Machine learning; Posture analysis; Sleep-like state.

PubMed Disclaimer

Conflict of interest statement

Competing interests The authors declare no competing or financial interests.

Figures

Fig. 1.
Fig. 1.
A schematic of the cricket behaviour-recording system. (A) An overview of the quantitative evaluation system for recording, through which we simultaneously recorded up to six crickets in one experiment by isolating individuals using acrylic cases (150×100×50 mm). The 12-h light/dark cycle (LD 12:12) was controlled by an electric time switch. The movement of crickets was recorded by a video camera, set at a height of 450 mm above the floor and connected to a Raspberry Pi 3. (B) A photograph depicting the specific recording environment. Sheets of paper with a height of 15 mm were placed between each acrylic case to make the individuals invisible to each other. (C) A schematic diagram of the acrylic case. The crickets were allowed free access to water and food. The water source is the tube-stuffed cotton, and the feeding area consists of a section of the plastic case with holes.
Fig. 2.
Fig. 2.
Outline view of the system. Scheme from image acquisition to behavioural quantification. First, image data were recorded with a camera module connected Raspberry Pi at an interval of one image per minute. Images were converted into videos, and a machine learning algorithm, ResNet-50 and DeepLabCut, was employed for training and refinement to construct a model with six automatically labelled keypoints on crickets’ head, thorax, abdomen, abdominal tip, left hind leg, and right hind leg. Subsequently, the coordinates of these labelled keypoints were extracted. After fixing errors, behaviours such as locomotion, feeding activity, and SLS were defined based on the coordinates and behaviours quantitatively analysed across all frames over a period of 14 days.
Fig. 3.
Fig. 3.
A quantitative analysis of locomotion of crickets. (A) The double-plotted actogram recorded under a 12-h light:dark (LD) cycle at 30°C for 14 days shows that the frequency of locomotion increases during the dark phase. (B,C) Mean number of frames of locomotion for 12 h (B) and 1 h (C) indicated that locomotion significantly increases during the dark phase (1 frame per min).
Fig. 4.
Fig. 4.
A quantitative analysis of feeding activity of crickets. (A,B) Mean number of frames of feeding activity for 12 h (A) and 1 h (B) suggested that peaks were evident around the transition from dark to light, though there was no significant difference in feeding activity between the dark and light phases (1 frame per min).
Fig. 5.
Fig. 5.
A quantitative analysis of SLS of crickets. (A) The double-plotted actogram recorded under a 12-h light:dark (LD) cycle at 30°C for 14 days shows that the frequency of SLS increases during the light phase. (B,C) Mean number of frames of SLS for 12 h (B) and 1 h (C) indicated that the SLS significantly increases during the light phase, with a second peak in the late dark phase (1 frame per min).
Fig. 6.
Fig. 6.
Body posture associated with the SLS in crickets. Body posture and leg relaxation were calculated from the angle between three points: the abdomen, left hind leg, and right hind leg. The relationship with the SLS was then examined. (A,B) The mean angle for 12 h (A) and 1 h (B) indicated that leg relaxation was significantly increased during the light period, when the frequency of the SLS was higher. (C) A scatter plot of the mean angle per hour against the mean SLS frequency corresponding to same hour was calculated with a fitted linear regression line. The equation of the line, with R squared as the coefficient of determination and P-value, indicates a negative correlation between hind legs angle and the number of continuous immobility frames. (D) Mean hind legs angle was calculated for each immobile time from 0 to 20 min with a fitted linear regression line. The equation of the line, with R squared as the coefficient of determination and P-value, indicates a strong negative correlation.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Abe, Y., Ushirogawa, H. and Tomioka, K. (1997). Circadian locomotor rhythms in the cricket, Gryllodes sigillatus I. Localization of the pacemaker and the photoreceptor. Zool. Sci. 14, 719-727. 10.2108/zsj.14.719 - DOI - PubMed
    1. Ajayi, O. M., Marlman, J. M., Gleitz, L. A., Smith, E. S., Piller, B. D., Krupa, J. A., Vinauger, C. and Benoit, J. B. (2022). Behavioral and postural analyses establish sleep-like states for mosquitoes that can impact host landing and blood feeding. J. Exp. Biol. 225, jeb244032. 10.1242/jeb.244032 - DOI - PMC - PubMed
    1. Allada, R., Cirelli, C. and Sehgal, A. (2017). Molecular mechanisms of sleep homeostasis in flies and mammals. Cold Spring Harb. Perspect. Biol. 9, 1-20. 10.1101/cshperspect.a027730 - DOI - PMC - PubMed
    1. Anderson, D. J. and Perona, P. (2014). Toward a science of computational ethology. Neuron 84, 18-31. 10.1016/j.neuron.2014.09.005 - DOI - PubMed
    1. Blankers, T., Oh, K. P., Bombarely, A. and Shaw, K. L. (2018). The genomic architecture of a rapid island radiation: recombination rate variation, chromosome structure, and genome assembly of the Hawaiian Cricket Laupala. Genetics 209, 1329-1344. 10.1534/genetics.118.300894 - DOI - PMC - PubMed

Publication types

-