Orientation-invariant autoencoders learn robust representations for shape profiling of cells and organelles.

Autor: Burgess J; Institute for Computational & Mathematical Engineering, Stanford University, Stanford, CA, USA. jmhb@stanford.edu., Nirschl JJ; Department of Pathology, School of Medicine, Stanford University, Stanford, CA, USA., Zanellati MC; Department of Cell Biology and Physiology, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA., Lozano A; Department of Biomedical Data Science, Stanford University, Stanford, CA, USA., Cohen S; Department of Cell Biology and Physiology, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA., Yeung-Levy S; Departments of Biomedical Data Science, Computer Science, and Electrical Engineering, Stanford University, Stanford, CA, USA. syyeung@stanford.edu.; Chan Zuckerberg Biohub - San Francisco, San Francisco, CA, USA. syyeung@stanford.edu.; Clinical Excellence Research Center, School of Medicine, Stanford University, Stanford, CA, USA. syyeung@stanford.edu.
Jazyk: angličtina
Zdroj: Nature communications [Nat Commun] 2024 Feb 03; Vol. 15 (1), pp. 1022. Date of Electronic Publication: 2024 Feb 03.
DOI: 10.1038/s41467-024-45362-4
Abstrakt: Cell and organelle shape are driven by diverse genetic and environmental factors and thus accurate quantification of cellular morphology is essential to experimental cell biology. Autoencoders are a popular tool for unsupervised biological image analysis because they learn a low-dimensional representation that maps images to feature vectors to generate a semantically meaningful embedding space of morphological variation. The learned feature vectors can also be used for clustering, dimensionality reduction, outlier detection, and supervised learning problems. Shape properties do not change with orientation, and thus we argue that representation learning methods should encode this orientation invariance. We show that conventional autoencoders are sensitive to orientation, which can lead to suboptimal performance on downstream tasks. To address this, we develop O2-variational autoencoder (O2-VAE), an unsupervised method that learns robust, orientation-invariant representations. We use O2-VAE to discover morphology subgroups in segmented cells and mitochondria, detect outlier cells, and rapidly characterise cellular shape and texture in large datasets, including in a newly generated synthetic benchmark.
(© 2024. The Author(s).)
Databáze: MEDLINE
Externí odkaz: