EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

PhaseXplorer Creates High-Dimensional Phase Diagrams with Closed-Loop Active Learning

Stef A H Jansen iD, Lasse S A Dreyer, Jule van Basten, Yihan Yao, Daniel E Otzen iD, Tom F A de Greef iD, Tuomas P J Knowles iD, E W Meijer iD, Nadia A Erkamp iD

ACS Nano · 18.11.2025 · 19 (45) · 38981-38989

DOI: 10.1021/acsnano.5c07268 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACS Nano, online erschienen am 03.11.2025. Verfasst von Stef A H Jansen und 8 weiteren Autor:innen. Beteiligte Einrichtung: Institute for Complex Molecular Systems (ICMS), Eindhoven University of Technology, Eindhoven 5612 AJ, The Netherlands. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Innovationsfonden, Ministerie van Onderwijs, Cultuur en Wetenschap, China Scholarship Council.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Phase separation fundamentally governs material properties and cellular function on multiple organizational scales. Conventional approaches to studying this nonlinear phenomenon necessitate resource-intensive experiments. As such, investigations were limited to low-dimensional space. We present PhaseXplorer, a platform that combines microfluidics, microscopy, and machine learning to efficiently study phase separation systems. PhaseXplorer autonomously designs, generates, and analyzes samples in a closed-loop active learning workflow until an accurate phase diagram is obtained. Using an acquisition function that balances exploration and exploitation, all the phase boundaries are located with minimal sampling. A convolutional neural network executes real-time image recognition to identify microfluidic droplets and phase separation within them in less than 1 ms per droplet. PhaseXplorer standardizes analysis across experiments and does not require calibration nor extensive postexperiment analysis. We demonstrate PhaseXplorer's capabilities using a poly rA model system by creating a four-dimensional phase diagram 100 times faster than traditional methods while simultaneously consuming 10,000 times less material.

Schlagwörter

Stichwörter der Autor:innen

PhaseXploreractive learningclosed-loopmicrofluidicsphase diagramphase separation

Autor:innen und Institutionen

  • Institute for Complex Molecular Systems (ICMS), Eindhoven University of Technology, Eindhoven 5612 AJ, The Netherlands.
  • Interdisciplinary Nanoscience Center (iNANO), Aarhus University, Aarhus 8000, Denmark.
  • Yusuf Hamied Department of Chemistry, University of Cambridge, Cambridge CB2 1EW, U.K.
  • Cavendish Laboratory, Department of Physics, University of Cambridge, Cambridge CB3 0HE, U.K.
  • School of Chemistry and RNA Institute, University of New South Wales, Sydney 2052, Australia.

ORCID-Profile: Stef A H Jansen, Daniel E Otzen, Tom F A de Greef, Tuomas P J Knowles, E W Meijer, Nadia A Erkamp

Förderung

  • Innovationsfonden
  • Ministerie van Onderwijs, Cultuur en Wetenschap
  • China Scholarship Council
  • Cambridge Trust
  • H2020 European Research Council
  • European Molecular Biology Organization

Zitieren

APA

Jansen, S. A. H., Dreyer, L. S. A., van Basten, J., Yao, Y., Otzen, D. E., de Greef, T. F. A., Knowles, T. P. J., Meijer, E. W., & Erkamp, N. A. (2025). PhaseXplorer Creates High-Dimensional Phase Diagrams with Closed-Loop Active Learning. ACS Nano, 19(45), 38981-38989. https://doi.org/10.1021/acsnano.5c07268

Vancouver

Jansen SAH, Dreyer LSA, van Basten J, Yao Y, Otzen DE, de Greef TFA, et al. PhaseXplorer Creates High-Dimensional Phase Diagrams with Closed-Loop Active Learning. ACS Nano. 2025;19(45):38981-38989. doi:10.1021/acsnano.5c07268

BibTeX
@article{Jansen2025,
  author = {Jansen, Stef A H and Dreyer, Lasse S A and van Basten, Jule and Yao, Yihan and Otzen, Daniel E and de Greef, Tom F A and Knowles, Tuomas P J and Meijer, E W and Erkamp, Nadia A},
  title = {PhaseXplorer Creates High-Dimensional Phase Diagrams with Closed-Loop Active Learning},
  journal = {ACS Nano},
  year = {2025},
  volume = {19},
  number = {45},
  pages = {38981--38989},
  doi = {10.1021/acsnano.5c07268}
}
Kostenlos starten →