idw – Informationsdienst Wissenschaft
Nachrichten, Termine, Experten
Nachrichten, Termine, Experten
idw - Informationsdienst
Wissenschaft
A joint team of researchers from Constructor University and Constructor Technology have harnessed the power of machine learning to uncover new findings about human thyroid health. In a new study published in Frontiers in Endocrinology, a team led by Constructor University Professor of Cell Biology Klaudia Brix developed a new image-analysis app called “CU Cilia,” which leveraged machine learning algorithms to successfully detect and measure primary cilia in human thyroid cells.
In addition to showcasing CU Cilia’s improved speed, precision and accessibility compared to conventional methods, the study confirmed a link between certain enzymes and primary cilia length in thyroid cells, a finding that could carry important implications for thyroid health.
Primary cilia are tiny, hair-like projections that protrude from the cell membrane. The thyroid epithelial cells analyzed in the study provide the body with vital thyroid hormones, using primary cilia as a sort of antenna to sense environmental signals and regulate these crucial processes. The research project looked at whether primary cilia in human thyroid cells are affected by enzymes known as cathepsins, which earlier research has linked to cilia development and thyroid health in mice. As Professor Brix noted, “with this study, we wanted to investigate whether the structure of primary cilia depends on proteases,” the class of enzymes that include cathepsins and break down proteins and peptides. “This is an important basic science question because these delicate cellular antennas serve as biomarkers of health and disease.”
The project entailed analyzing large numbers of high-resolution, laser-scanning microscopy images of primary cilia, which is where Professor Brix saw an opportunity to innovate and collaborate with machine-learning experts from the university’s partner organization Constructor Technology. “In a nutshell, CU Cilia allows us to analyze hundreds of images, making the approach of using structural alterations of primary cilia as health and disease biomarkers accessible to scientists of basic science and medical research fields.”
Leveraging the interdisciplinary expertise of the two Constructor Group organizations, the project team tested CU Cilia alongside conventional rule-based image analysis programs and succeeded at delivering faster and more accessible analysis for detecting primary cilia and segmenting nuclei. “After we learned to talk to each other about the underlying cell biology and the computational challenges of high-content image analysis, it was a great experience to work on realizing this machine learning-based app,” reflected Research Assistant Maren Rehders.
Analysis using CU Cilia confirmed a link between cathepsins and primary cilia length, suggesting a structural role for the enzyme in thyroid cells for both humans and mice, and opening new pathways for further investigation. By combining the precision of machine learning with the biological insight of high-resolution microscopy, the study demonstrated the potential of machine learning to accelerate discovery—even at the smallest scales of life. "This work highlights the power of interdisciplinary collaboration, where machine learning engineers and molecular biologists can join forces to drive scientific discovery and create impactful tools,” added Constructor Group machine-learning expert Professor Peter Popov.
Dr. Klaudia Brix, kbrix@constructor.university
https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2025.1...
Prof. Dr. Klaudia Brix, Professorin für Zellbiologie an der Constructor University
Merkmale dieser Pressemitteilung:
Journalisten, Wissenschaftler, jedermann
Biologie, Chemie, Ernährung / Gesundheit / Pflege, Medizin
überregional
Buntes aus der Wissenschaft, Forschungsergebnisse
Englisch
Sie können Suchbegriffe mit und, oder und / oder nicht verknüpfen, z. B. Philo nicht logie.
Verknüpfungen können Sie mit Klammern voneinander trennen, z. B. (Philo nicht logie) oder (Psycho und logie).
Zusammenhängende Worte werden als Wortgruppe gesucht, wenn Sie sie in Anführungsstriche setzen, z. B. „Bundesrepublik Deutschland“.
Die Erweiterte Suche können Sie auch nutzen, ohne Suchbegriffe einzugeben. Sie orientiert sich dann an den Kriterien, die Sie ausgewählt haben (z. B. nach dem Land oder dem Sachgebiet).
Haben Sie in einer Kategorie kein Kriterium ausgewählt, wird die gesamte Kategorie durchsucht (z.B. alle Sachgebiete oder alle Länder).
