Accurate and versatile 3D segmentation of plant tissues at cellular resolution
Abstract
Quantitative analysis of plant and animal morphogenesis requires accurate segmentation of individual cells in volumetric images of growing organs. In the last years, deep learning has provided robust automated algorithms that approach human performance, with applications to bio-image analysis now starting to emerge. Here, we present PlantSeg, a pipeline for volumetric segmentation of plant tissues into cells. PlantSeg employs a convolutional neural network to predict cell boundaries and graph partitioning to segment cells based on the neural network predictions. PlantSeg was trained on 1xed and live plant organs imaged with confocal and light sheet microscopes. PlantSeg delivers accurate results and generalizes well across different tissues, scales, acquisition settings even on non plant samples. We present results of PlantSeg applications in diverse developmental contexts. PlantSeg is free and open-source, with both a command line and a user-friendly graphical interface (https://github.com/hci-unihd/plant-seg).
Data availability
All data used in this study have been deposited in Open Science Framework: https://osf.io/uzq3w/Additionally Arabidopsis 3D Digital Tissue Atlas is available under https://osf.io/fzr56/
Article and author information
Author details
Funding
Deutsche Forschungsgemeinschaft (FOR2581)
- Jan U Lohmann
- Miltos Tsiantis
- Fred Hamprecht
- Kay Schneitz
- Alexis Maizel
- Anna Kreshuk
Leverhulme Trust (RPG-2016-049)
- George Bassel
The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.
Copyright
© 2020, Wolny et al.
This article is distributed under the terms of the Creative Commons Attribution License permitting unrestricted use and redistribution provided that the original author and source are credited.
Metrics
-
- 12,934
- views
-
- 1,500
- downloads
-
- 185
- citations
Views, downloads and citations are aggregated across all versions of this paper published by eLife.
Download links
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Further reading
-
- Microbiology and Infectious Disease
- Plant Biology
Programmed cell death occurring during plant development (dPCD) is a fundamental process integral for plant growth and reproduction. Here, we investigate the connection between developmentally controlled PCD and fungal accommodation in Arabidopsis thaliana roots, focusing on the root cap-specific transcription factor ANAC033/SOMBRERO (SMB) and the senescence-associated nuclease BFN1. Mutations of both dPCD regulators increase colonization by the beneficial fungus Serendipita indica, primarily in the differentiation zone. smb-3 mutants additionally exhibit hypercolonization around the meristematic zone and a delay of S. indica-induced root-growth promotion. This demonstrates that root cap dPCD and rapid post-mortem clearance of cellular corpses represent a physical defense mechanism restricting microbial invasion of the root. Additionally, reporter lines and transcriptional analysis revealed that BFN1 expression is downregulated during S. indica colonization in mature root epidermal cells, suggesting a transcriptional control mechanism that facilitates the accommodation of beneficial microbes in the roots.
-
- Cell Biology
- Plant Biology
Plants distribute many nutrients to chloroplasts during leaf development and maturation. When leaves senesce or experience sugar starvation, the autophagy machinery degrades chloroplast proteins to facilitate efficient nutrient reuse. Here, we report on the intracellular dynamics of an autophagy pathway responsible for piecemeal degradation of chloroplast components. Through live-cell monitoring of chloroplast morphology, we observed the formation of chloroplast budding structures in sugar-starved leaves. These buds were then released and incorporated into the vacuolar lumen as an autophagic cargo termed a Rubisco-containing body. The budding structures did not accumulate in mutants of core autophagy machinery, suggesting that autophagosome creation is required for forming chloroplast buds. Simultaneous tracking of chloroplast morphology and autophagosome development revealed that the isolation membranes of autophagosomes interact closely with part of the chloroplast surface before forming chloroplast buds. Chloroplasts then protrude at the site associated with the isolation membranes, which divide synchronously with autophagosome maturation. This autophagy-related division does not require DYNAMIN-RELATED PROTEIN 5B, which constitutes the division ring for chloroplast proliferation in growing leaves. An unidentified division machinery may thus fragment chloroplasts for degradation in coordination with the development of the chloroplast-associated isolation membrane.