DeepFly3D, a deep learning-based approach for 3D limb and appendage tracking in tethered, adult Drosophila

Abstract

Studying how neural circuits orchestrate limbed behaviors requires the precise measurement of the positions of each appendage in 3-dimensional (3D) space. Deep neural networks can estimate 2-dimensional (2D) pose in freely behaving and tethered animals. However, the unique challenges associated with transforming these 2D measurements into reliable and precise 3D poses have not been addressed for small animals including the fly, Drosophila melanogaster. Here we present DeepFly3D, a software that infers the 3D pose of tethered, adult Drosophila using multiple camera images. DeepFly3D does not require manual calibration, uses pictorial structures to automatically detect and correct pose estimation errors, and uses active learning to iteratively improve performance. We demonstrate more accurate unsupervised behavioral embedding using 3D joint angles rather than commonly used 2D pose data. Thus, DeepFly3D enables the automated acquisition of Drosophila behavioral measurements at an unprecedented level of detail for a variety of biological applications.

Data availability

All data generated and analyzed during this study are included in the DeepFly3D GitHub site: https://github.com/NeLy-EPFL/DeepFly3D and in the Harvard Dataverse.

The following data sets were generated

Article and author information

Author details

  1. Semih Günel

    School of Computer and Communication Sciences, Computer Vision Laboratory, EPFL, Lausanne, Switzerland
    For correspondence
    semih.gunel@epfl.ch
    Competing interests
    The authors declare that no competing interests exist.
  2. Helge Rhodin

    School of Computer and Communication Sciences, Computer Vision Laboratory, EPFL, Lausanne, Switzerland
    Competing interests
    The authors declare that no competing interests exist.
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-2692-0801
  3. Daniel Morales

    School of Life Sciences, Brain Mind Institute and Interfaculty Institute of Bioengineering, Neuroengineering Laboratory, EPFL, Lausanne, Switzerland
    Competing interests
    The authors declare that no competing interests exist.
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-7469-0898
  4. João H Campagnolo

    School of Life Sciences, Brain Mind Institute and Interfaculty Institute of Bioengineering, Neuroengineering Laboratory, EPFL, Lausanne, Switzerland
    Competing interests
    The authors declare that no competing interests exist.
  5. Pavan Ramdya

    School of Life Sciences, Brain Mind Institute and Interfaculty Institute of Bioengineering, Neuroengineering Laboratory, EPFL, Lausanne, Switzerland
    For correspondence
    pavan.ramdya@epfl.ch
    Competing interests
    The authors declare that no competing interests exist.
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-5425-4610
  6. Pascal Fua

    School of Computer and Communication Sciences, Computer Vision Laboratory, EPFL, Lausanne, Switzerland
    Competing interests
    The authors declare that no competing interests exist.

Funding

Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung (175667)

  • Daniel Morales
  • Pavan Ramdya

Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung (181239)

  • Daniel Morales
  • Pavan Ramdya

EPFL (iPhD)

  • Semih Günel

Microsoft Research (JRC Project)

  • Helge Rhodin

Swiss Government Excellence Postdoctoral Scholarship (2018.0483)

  • Daniel Morales

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Copyright

© 2019, Günel et al.

This article is distributed under the terms of the Creative Commons Attribution License permitting unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 8,242
    views
  • 900
    downloads
  • 133
    citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Semih Günel
  2. Helge Rhodin
  3. Daniel Morales
  4. João H Campagnolo
  5. Pavan Ramdya
  6. Pascal Fua
(2019)
DeepFly3D, a deep learning-based approach for 3D limb and appendage tracking in tethered, adult Drosophila
eLife 8:e48571.
https://doi.org/10.7554/eLife.48571

Share this article

https://doi.org/10.7554/eLife.48571

Further reading

    1. Developmental Biology
    2. Neuroscience
    Ev L Nichols, Joo Lee, Kang Shen
    Research Article

    During development axons undergo long-distance migrations as instructed by guidance molecules and their receptors, such as UNC-6/Netrin and UNC-40/DCC. Guidance cues act through long-range diffusive gradients (chemotaxis) or local adhesion (haptotaxis). However, how these discrete modes of action guide axons in vivo is poorly understood. Using time-lapse imaging of axon guidance in C. elegans, we demonstrate that UNC-6 and UNC-40 are required for local adhesion to an intermediate target and subsequent directional growth. Exogenous membrane-tethered UNC-6 is sufficient to mediate adhesion but not directional growth, demonstrating the separability of haptotaxis and chemotaxis. This conclusion is further supported by the endogenous UNC-6 distribution along the axon’s route. The intermediate and final targets are enriched in UNC-6 and separated by a ventrodorsal UNC-6 gradient. Continuous growth through the gradient requires UNC-40, which recruits UNC-6 to the growth cone tip. Overall, these data suggest that UNC-6 stimulates stepwise haptotaxis and chemotaxis in vivo.

    1. Neuroscience
    Mihály Vöröslakos, Yunchang Zhang ... György Buzsáki
    Tools and Resources

    Brain states fluctuate between exploratory and consummatory phases of behavior. These state changes affect both internal computation and the organism’s responses to sensory inputs. Understanding neuronal mechanisms supporting exploratory and consummatory states and their switching requires experimental control of behavioral shifts and collecting sufficient amounts of brain data. To achieve this goal, we developed the ThermoMaze, which exploits the animal’s natural warmth-seeking homeostatic behavior. By decreasing the floor temperature and selectively heating unmarked areas, we observed that mice avoided the aversive state by exploring the maze and finding the warm spot. In its design, the ThermoMaze is analogous to the widely used water maze but without the inconvenience of a wet environment and, therefore, allows the collection of physiological data in many trials. We combined the ThermoMaze with electrophysiology recording, and report that spiking activity of hippocampal CA1 neurons during sharp-wave ripple events encode the position of mice. Thus, place-specific firing is not confined to locomotion and associated theta oscillations but persist during waking immobility and sleep at the same location. The ThermoMaze will allow for detailed studies of brain correlates of immobility, preparatory–consummatory transitions, and open new options for studying behavior-mediated temperature homeostasis.