Accurate and versatile 3D segmentation of plant tissues at cellular resolution

Wolny, Adrian; Cerrone, Lorenzo; Vijayan, Athul; Tofanelli, Rachele; Barro, Amaya Vilches; Louveaux, Marion; Wenzl, Christian; Strauss, Soeren; Wilson-Sánchez, David; Lymbouridou, Rena; Steigleder, Susanne S; Pape, Constantin; Bailoni, Alberto; Duran-Nebreda, Salva; Bassel, George W.; Lohmann, Jan U.; Tsiantis, Miltos; Hamprecht, Fred A.; Schneitz, Kay; Maizel, Alexis; Kreshuk, Anna

doi:10.7554/elife.57613

Cited by 197 publications

(209 citation statements)

References 61 publications

Supporting

Mentioning

209

Contrasting

Order By: Relevance

“…We processed the raw 3D datasets using PlantSeg, a deep learning pipeline for 3D segmentation of dense plant tissue at cellular resolution (Wolny et al, 2020) (Fig. 1E).…”

Section: Resultsmentioning

confidence: 99%

“…The 3D digital ovules including their quantitative descriptions can be downloaded from the BioStudies data repository at EMBL-EBI (Sarkans et al, 2018). The experimental strategy is applicable to other organs as well (Tofanelli et al, 2019); Wolny et al, 2020). In an exemplary manner, this work demonstrates the exciting new insights that can be obtained when studying tissue morphogenesis in full 3D.…”

Section: Discussionmentioning

confidence: 99%

“…3D cells were segmented using PlantSeg (Wolny et al 2020). Cell segmentation using boundary predictions in PlantSeg allowed us to process microscopic image z-stacks captured with coarse voxels, to minimise the segmentation errors, and to reduce the imaging time and hence laser exposure and phototoxicity to samples.…”

Section: Methodsmentioning

confidence: 99%

See 2 more Smart Citations

A digital 3D reference atlas reveals cellular growth patterns shaping the Arabidopsis ovule

Vijayan

Tofanelli

Strauss

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

A fundamental question in biology is how morphogenesis integrates the multitude of distinct processes that act at different scales, ranging from the molecular control of gene expression to cellular coordination in a tissue. Investigating morphogenesis of complex organs strongly benefits from three-dimensional representations of the organ under study. Here, we present a digital analysis of ovule development from Arabidopsis thaliana as a paradigm for a complex morphogenetic process. Using machine-learning-based image analysis we generated a three-dimensional atlas of ovule development with cellular resolution. It allows quantitative stage- and tissue-specific analysis of cellular patterns. Exploiting a fluorescent reporter enabled precise spatial determination of gene expression patterns, revealing subepidermal expression of WUSCHEL. Underlying the power of our approach, we found that primordium outgrowth progresses evenly, discovered a novel mode of forming a new cell layer, and detected a new function of INNER NO OUTER in restricting cell proliferation in the nucellus. Moreover, we identified two distinct subepidermal cell populations that make crucial contributions to ovule curvature. Our work demonstrates the expedience of a three-dimensional digital representation when studying the morphogenesis of an organ of complex cellular architecture and shape that eventually consists of 1,900 cells.

show abstract

“…We processed the raw 3D datasets using PlantSeg, a deep learning pipeline for 3D segmentation of dense plant tissue at cellular resolution (Wolny et al, 2020) (Fig. 1E).…”

Section: Resultsmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

A digital 3D reference atlas reveals cellular growth patterns shaping the Arabidopsis ovule

Vijayan

Tofanelli

Strauss

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…They can be found linked to dispersed ACDs, in contiguous cell files aligned with organ axes, or nestled in cell corners. POME can quantify dynamic polarity in these different contexts and complements other recent advances in cell segmentation and quantitative image analysis (Wolny et al, 2020). POME converts a once low-throughput and qualitative procedure into a semi-automated one that is amicable to statistical analysis.…”

Section: Resultsmentioning

confidence: 91%

Quantitative and dynamic cell polarity tracking in plant cells

Gong

Varnau

Wallner

et al. 2020

Preprint

View full text Add to dashboard Cite

Quantitative information on the spatiotemporal distribution of polarized proteins is central for understanding cell-fate determination, yet collecting sufficient data for statistical analysis is difficult to accomplish with manual measurements. Here we present POME, a semi-automated pipeline for the quantification of cell polarity, and demonstrate its application to a variety of developmental contexts. POME analysis reveals that during asymmetric cell divisions in the Arabidopsis thaliana stomatal lineage, polarity proteins BASL and BRXL2 are more asynchronous and less mutually dependent than previously thought. While their interaction is important to maintain their polar localization and recruit other effectors to regulate asymmetric cell divisions, BRXL2 polarization precedes that of BASL and can be initiated in BASL's absence. Uncoupling of polarization from BASL activity is also seen in Brachypodium distachyon, where we find that the MAPKKK BdYDA1 is segregated and polarized following asymmetric division. Our results demonstrate that POME is a versatile tool, which by itself or combined with tissue-level studies and advanced microscopy techniques can help uncover new mechanisms of cell polarity.

show abstract

“…Next, we sought to expand the capability of OpSeF to volume segmentation. To this aim, we trained a StarDist 3D model using the annotation of Arabidopsis thaliana lateral root nuclei dataset provided by Wolny et al (2020). Images were obtained on a Luxendo MuVi SPIM light-sheet microscope (Wolny et al, 2020).…”

Section: Resultsmentioning

confidence: 99%

OpSeF: Open Source Python Framework for Collaborative Instance Segmentation of Bioimages

Rasse

Hollandi

Horváth

2020

Front. Bioeng. Biotechnol.

View full text Add to dashboard Cite

Various pre-trained deep learning models for the segmentation of bioimages have been made available as developer-to-end-user solutions. They are optimized for ease of use and usually require neither knowledge of machine learning nor coding skills. However, individually testing these tools is tedious and success is uncertain. Here, we present the Open Segmentation Framework (OpSeF), a Python framework for deep learningbased instance segmentation. OpSeF aims at facilitating the collaboration of biomedical users with experienced image analysts. It builds on the analysts' knowledge in Python, machine learning, and workflow design to solve complex analysis tasks at any scale in a reproducible, well-documented way. OpSeF defines standard inputs and outputs, thereby facilitating modular workflow design and interoperability with other software. Users play an important role in problem definition, quality control, and manual refinement of results. OpSeF semi-automates preprocessing, convolutional neural network (CNN)based segmentation in 2D or 3D, and postprocessing. It facilitates benchmarking of multiple models in parallel. OpSeF streamlines the optimization of parameters for preand postprocessing such, that an available model may frequently be used without retraining. Even if sufficiently good results are not achievable with this approach, intermediate results can inform the analysts in the selection of the most promising CNNarchitecture in which the biomedical user might invest the effort of manually labeling training data. We provide Jupyter notebooks that document sample workflows based on various image collections. Analysts may find these notebooks useful to illustrate common segmentation challenges, as they prepare the advanced user for gradually taking over some of their tasks and completing their projects independently. The notebooks may also be used to explore the analysis options available within OpSeF in an interactive way and to document and share final workflows. Currently, three mechanistically distinct CNN-based segmentation methods, the U-Net implementation used in Cellprofiler 3.0, StarDist, and Cellpose have been integrated within OpSeF. The addition of new networks requires little; the addition of new models requires no coding skills. Thus, OpSeF might soon become both an interactive model repository, in which pre-trained models might be shared, evaluated, and reused with ease.

show abstract

Accurate and versatile 3D segmentation of plant tissues at cellular resolution

Cited by 197 publications

References 61 publications

A digital 3D reference atlas reveals cellular growth patterns shaping the Arabidopsis ovule

A digital 3D reference atlas reveals cellular growth patterns shaping the Arabidopsis ovule

Quantitative and dynamic cell polarity tracking in plant cells

OpSeF: Open Source Python Framework for Collaborative Instance Segmentation of Bioimages

Contact Info

Product

Resources

About