Revealing invisible cell phenotypes with conditional generative modeling.

Autor: Lamiable A; Computational Bioimaging and Bioinformatics, Institut de Biologie de l'Ecole Normale Supérieure, PSL University, 46, rue d'Ulm, 75005, Paris, France., Champetier T; Computational Bioimaging and Bioinformatics, Institut de Biologie de l'Ecole Normale Supérieure, PSL University, 46, rue d'Ulm, 75005, Paris, France.; Ksilink, 16 rue d'Ankara, 67000, Strasbourg, France., Leonardi F; Computational Bioimaging and Bioinformatics, Institut de Biologie de l'Ecole Normale Supérieure, PSL University, 46, rue d'Ulm, 75005, Paris, France.; Université Paris-Cité, MERIT, IRD, F-75006, Paris, France., Cohen E; Computational Bioimaging and Bioinformatics, Institut de Biologie de l'Ecole Normale Supérieure, PSL University, 46, rue d'Ulm, 75005, Paris, France., Sommer P; Ksilink, 16 rue d'Ankara, 67000, Strasbourg, France., Hardy D; Histopathology Platform, Institut Pasteur, F-75015, Paris, France., Argy N; Université Paris-Cité, MERIT, IRD, F-75006, Paris, France.; Laboratoire de parasitologie-mycologie, Hôpital Bichat-Claude bernard, APHP, Paris, France., Massougbodji A; Institut de Recherche Clinique du Bénin, Abomey-Calavi, Benin., Del Nery E; Biophenics, Institut Curie, PSL Research University, Department of Translational Research, Cell and Tissue Imaging Facility (PICT-IBiSA), 26 rue d'Ulm, 75005, Paris, France., Cottrell G; Université Paris-Cité, MERIT, IRD, F-75006, Paris, France., Kwon YJ; Ksilink, 16 rue d'Ankara, 67000, Strasbourg, France. yong-jun.kwon@lih.lu.; Personalized Therapy Discovery, Department of Oncology, Luxembourg Institute of Health, Dudelange, Luxembourg. yong-jun.kwon@lih.lu., Genovesio A; Computational Bioimaging and Bioinformatics, Institut de Biologie de l'Ecole Normale Supérieure, PSL University, 46, rue d'Ulm, 75005, Paris, France. auguste.genovesio@ens.psl.eu.
Jazyk: angličtina
Zdroj: Nature communications [Nat Commun] 2023 Oct 11; Vol. 14 (1), pp. 6386. Date of Electronic Publication: 2023 Oct 11.
DOI: 10.1038/s41467-023-42124-6
Abstrakt: Biological sciences, drug discovery and medicine rely heavily on cell phenotype perturbation and microscope observation. However, most cellular phenotypic changes are subtle and thus hidden from us by natural cell variability: two cells in the same condition already look different. In this study, we show that conditional generative models can be used to transform an image of cells from any one condition to another, thus canceling cell variability. We visually and quantitatively validate that the principle of synthetic cell perturbation works on discernible cases. We then illustrate its effectiveness in displaying otherwise invisible cell phenotypes triggered by blood cells under parasite infection, or by the presence of a disease-causing pathological mutation in differentiated neurons derived from iPSCs, or by low concentration drug treatments. The proposed approach, easy to use and robust, opens the door to more accessible discovery of biological and disease biomarkers.
(© 2023. Springer Nature Limited.)
Databáze: MEDLINE
Externí odkaz: