Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Correspondence
  • Published:

DL4MicEverywhere: deep learning for microscopy made flexible, shareable and reproducible

Nature Methods (2024)Cite this article

Subjects

DL4MicEverywhere is a platform that lets users train and implement their models in different computational environments. These environments include Google Colab, personal computational resources such as a desktop or laptop, and HPC systems. This flexibility is achieved by encapsulating each deep learning technique in an interactive Jupyter Notebook within a Docker container, enabling others to replicate analyses consistently across multiple platforms. DL4MicEverywhere (https://github.com/HenriquesLab/DL4MicEverywhere) enables users to install and interact with a large offering of standardized, user-friendly deep learning workflows, away from the limitations of proprietary platforms such as Google Colab and in a secure computational environment with controlled data privacy and resources. DL4MicEverywhere can be launched graphically, via X11 forwarding, or directly through a command line (headless mode), supporting HPC usage. This cross-platform containerization technology boosts the long-term platform’s sustainability and reproducibility, enhancing user convenience7.

DL4MicEverywhere features a zero-code interface that handles all the behind-the-scenes complexities, so users no longer need to deal with Docker configuration and deployment through a terminal. The intuitive interface abstracts away these technical details while providing a standardized Docker encapsulation for executing advanced techniques reliably. Researchers can select a notebook, choose computing resources and run the corresponding deep learning-powered analysis with just a few clicks (Fig. 1c–e). This allows users to train and apply models on various computing resources they control, eliminating reliance on third-party platforms. Furthermore, researchers can launch a notebook on local or remote systems with GPU acceleration whenever available, without worrying about complex software dependencies, Docker container management, or loss of access to deep learning frameworks (Fig. 1f–h). Compared to ZeroCostDL4Mic, DL4MicEverywhere doubles the number of deep learning approaches and provides new bioimaging analysis tasks, such as semantic segmentation, interactive instance segmentation, image registration, 3D single molecule localization microscopy, temporal and spatial upsampling, and image generation. The platform is designed to encourage the sharing and reuse of deep learning workflows provided as Jupyter Notebooks, which are then integrated into the BioImage Model Zoo. DL4MicEverywhere is strengthened by automated build pipelines8 that allow tracked versioning of ZeroCostDL4Mic notebooks and the seamless integration of new trainable models contributed by the community as user-friendly notebooks independently of the original ZeroCostDL4Mic framework (Fig. 1b). DL4MicEverywhere handles the corresponding testing and building of fully documented and open-source containers, making it easy for researchers to share not just the latest method, but the full software environment required to run it reliably.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: The DL4MicEverywhere platform.

Code availability

The source code, documentation and tutorials for DL4MicEverywhere are available at https://github.com/HenriquesLab/DL4MicEverywhere under a Creative Commons CC-BY-4.0 license.

References

  1. Moen, E. et al. Nat. Methods 16, 1233–1246 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. Pylvänäinen, J. W., Gómez-de-Mariscal, E., Henriques, R. & Jacquemet, G. Curr. Opin. Cell Biol. 85, 102271 (2023).

  3. Wei, O. et al. Preprint at bioRxiv https://doi.org/10.1101/2022.06.07.495102 (2022).

  4. Gómez-de-Mariscal, E. et al. Nat. Methods 18, 1192–1195 (2021).

  5. Laine, R. F., Arganda-Carreras, I., Henriques, R. & Jacquemet, G. Nat. Methods 18, 1136–1144 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. von Chamier, L. et al. Nat. Commun. 12, 2276 (2021).

    Article  Google Scholar 

  7. Moreau, D., Wiebels, K. & Boettiger, C. Nat. Rev. Methods Primers 3, 1–16 (2023).

    Article  Google Scholar 

  8. Beaulieu-Jones, B. K. & Greene, C. S. Nat. Biotechnol. 35, 342–346 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Bommasani, R. et al. Preprint at arXiv https://doi.org/10.48550/arXiv.2108.07258 (2022).

  10. Wang, H. et al. Nature 620, 47–60 (2023).

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

We thank Amin Rezaei, Ainhoa Serrano, Pablo Alonso, Urtzi Beorlegui, Andoni Rodriguez, Erlantz Calvo, Soham Mandal and Virginie Uhlmann for their contributions to the ZeroCostDL4Mic notebook collection. I.H.-C., M.G.F., C.T.R., R.H. and E.G.-d.-M. received funding from the European Union through the Horizon Europe program (AI4LIFE project with grant agreement 101057970-AI4LIFE and RT-SuperES project with grant agreement 101099654-RTSuperES to R.H.). I.H.-C., M.G.F., E.G.-d.-M. and R.H. also acknowledge the support of the Gulbenkian Foundation (Fundação Calouste Gulbenkian) and the European Research Council (ERC) under the European Union’s Horizon 2020 Research and Innovation Programme (grant agreement 101001332 to R.H.). Funded by the European Union. Views and opinions expressed are, however, those of the authors only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them. This work was also supported by a European Molecular Biology Organization (EMBO) installation grant (EMBO-2020-IG-4734 to R.H.), an EMBO postdoctoral fellowship (EMBO ALTF 174-2022 to E.G.-d.-M.), a Chan Zuckerberg Initiative Visual Proteomics Grant (vpi-0000000044 with https://doi.org/10.37921/743590vtudfp to R.H.). R.H. also acknowledges the support of LS4FUTURE Associated Laboratory (LA/P/0087/2020). This work is partially supported by grant GIU19/027 (to I.A.-C.) funded by the University of the Basque Country (UPV/EHU), grant PID2021-126701OB-I00 (to I.A.-C.) funded by the Ministerio de Ciencia, Innovación y Universidades, MICIU/AEI/10.13039/501100011033, and “ERDF A way of making Europe” (to I.A.-C.). This study was also supported by the Academy of Finland (338537 to G.J.), the Sigrid Juselius Foundation (to G.J.), the Cancer Society of Finland (Syöpäjärjestöt; to G.J.), and Solutions for Health strategic funding to Åbo Akademi University (to G.J.). This research was supported by the InFLAMES Flagship Programme of the Academy of Finland (decision number 337531).

Author information

Authors and Affiliations

  1. Optical Cell Biology Group, Instituto Gulbenkian de Ciência, Oeiras, Portugal

    Iván Hidalgo-Cenalmor, Mariana G. Ferreira, Mariana G. Ferreira, Ricardo Henriques & Estibaliz Gómez-de-Mariscal

  2. Faculty of Science and Engineering, Cell Biology, Åbo Akademi University, Turku, Finland

    Joanna W. Pylvänäinen & Guillaume Jacquemet

  3. European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Wellcome Genome Campus, Cambridge, UK

    Craig T. Russell, Matthew Hartley & Teresa Zulueta-Coarasa

  4. Department of Biomedical Engineering, Technion – Israel Institute of Technology, Haifa, Israel

    Alon Saguy & Yoav Shechtman

  5. Department of Computer Science and Artificial Intelligence, University of the Basque Country (UPV/EHU), San Sebastián, Spain

    Ignacio Arganda-Carreras

  6. IKERBASQUE, Basque Foundation for Science, Bilbao, Spain

    Ignacio Arganda-Carreras

  7. Donostia International Physics Center (DIPC), San Sebastián, Spain

    Ignacio Arganda-Carreras

  8. Biofisika Institute (CSIC-UPV/EHU), Leioa, Spain

    Ignacio Arganda-Carreras

  9. Department of Mechanical Engineering, University of Texas at Austin, Austin, TX, USA

    Yoav Shechtman

  10. Turku Bioscience Centre, University of Turku and Åbo Akademi University, Turku, Finland

    Guillaume Jacquemet

  11. Turku Bioimaging, University of Turku and Åbo Akademi University, Turku, Finland

    Guillaume Jacquemet

  12. InFLAMES Research Flagship Center, Åbo Akademi University, Turku, Finland

    Guillaume Jacquemet

  13. UCL Laboratory for Molecular Cell Biology, University College London, London, UK

    Ricardo Henriques

  14. Bioengineering Department, Universidad Carlos III de Madrid, Leganes, Spain

    Arrate Muñoz-Barrutia & Caterina Fuster Barcelo

  15. Instituto de Investigación Sanitaria Gregorio Marañón, Madrid, Spain

    Arrate Muñoz-Barrutia & Caterina Fuster Barcelo

  16. Euro-BioImaging ERIC Bio-Hub, European Molecular Biology Laboratory (EMBL) Heidelberg, Heidelberg, Germany

    Beatriz Serrano-Solano

  17. Georg-August-University Göttingen, Institute of Computer Science, Göttingen, Germany

    Constantin Pape

  18. Department of Pathology, Stanford University School of Medicine, Stanford, CA, USA

    Emma Lundberg

  19. Science for Life Laboratory, School of Engineering Sciences in Chemistry, Biotechnology and Health, KTH Royal Institute of Technology, Stockholm, Sweden

    Emma Lundberg

  20. Department of Bioengineering, Stanford University, Stanford, CA, USA

    Emma Lundberg

  21. Fondazione Human Technopole, Milan, Italy

    Florian Jug, Joran Deschamps, Mehdi Seifi & Vera Galinova

  22. Department of Applied Physics, Science for Life Laboratory, KTH Royal Institute of Technology, Stockholm, Sweden

    Wei Ouyang

Authors
  1. Iván Hidalgo-Cenalmor
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Joanna W. Pylvänäinen
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Mariana G. Ferreira
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Craig T. Russell
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Alon Saguy
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Ignacio Arganda-Carreras
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Yoav Shechtman
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Guillaume Jacquemet
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Ricardo Henriques
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Estibaliz Gómez-de-Mariscal
    View author publications

    You can also search for this author in PubMed Google Scholar

Consortia

AI4Life Horizon Europe Program Consortium

  • Arrate Muñoz-Barrutia
  • , Beatriz Serrano-Solano
  • , Caterina Fuster Barcelo
  • , Constantin Pape
  • , Craig T. Russell
  • , Emma Lundberg
  • , Estibaliz Gómez-de-Mariscal
  • , Florian Jug
  • , Joran Deschamps
  • , Iván Hidalgo-Cenalmor
  • , Mariana G. Ferreira
  • , Matthew Hartley
  • , Mehdi Seifi
  • , Ricardo Henriques
  • , Teresa Zulueta-Coarasa
  • , Vera Galinova
  •  & Wei Ouyang

Contributions

I.H.-C., G.J., R.H. and E.G.-d.-M. conceived, designed and wrote the source code of the project with contributions from all co-authors; I.H.-C., J.P.W., M.G.F., C.R., A.S., Y.S., G.J., R.H. and E.G.-d.-M. tested the platform; I.H.-C., J.P.W., M.G.F., G.J., R.H. and E.G.-d.-M. wrote the user documentation; I.H.-C., G.J., R.H. and E.G.-d.-M. wrote the paper with input from all co-authors. F.J. (florian.jug@fht.org) serves as a contact for the consortium.

Corresponding authors

Correspondence to Guillaume Jacquemet, Ricardo Henriques or Estibaliz Gómez-de-Mariscal.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Methods thanks Eugene Katrukha, Nils Körber and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Hidalgo-Cenalmor, I., Pylvänäinen, J.W., G. Ferreira, M. et al. DL4MicEverywhere: deep learning for microscopy made flexible, shareable and reproducible. Nat Methods (2024). https://doi.org/10.1038/s41592-024-02295-6

Download citation

  • Published:

  • DOI: https://doi.org/10.1038/s41592-024-02295-6

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing