A multiscale analysis of early flower development in Arabidopsis provides an integrated view of molecular regulation and growth control

Refahi, Yassin; Zardilis, Argyris; Michelin, Gaël; Wightman, Raymond; Leggio, Bruno; Legrand, Jonathan; Faure, Emmanuel; Vachez, Laetitia; Armezzani, Alessia; Risson, Anne Evodie; Zhao, Feng; Das, Pradeep; Prunet, Nathanaël; Meyerowitz, Elliot M.; Godin, Christophe; Malandain, Grégoire; Jönsson, Henrik; Traas, Jan

doi:10.1016/j.devcel.2021.01.019

Cited by 44 publications

(67 citation statements)

References 133 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For testing the segmentation algorithms, a dataset of ten 3D confocal stacks was used (described in (Refahi et al, 2021) ). First 6 stacks (0h, 24h, 32h, 72h, 120h, 132h timepoints) are from one meristem (FM1 in Refahi et al, 2021) while the time points 26h, 44h, 56h and 69h are from another (FM6). Images of the stacks are shown in Figure 4.…”

Section: Test Datasetmentioning

confidence: 99%

Benchmarking of deep learning algorithms for 3D instance segmentation of confocal image datasets

Kar

Petit

Refahi

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Segmenting three dimensional microscopy images is essential for understanding phenomena like morphogenesis, cell division, cellular growth and genetic expression patterns. Recently, deep learning (DL) pipelines have been developed which claim to provide high accuracy segmentation of cellular images and are increasingly considered as the state-of-the-art for image segmentation problems. However, it remains difficult to define their relative performance as the concurrent diversity and lack of uniform evaluation strategies makes it difficult to know how their results compare. In this paper, we first made an inventory of the available DL methods for 3D segmentation. We next implemented and quantitatively compared a number of representative DL pipelines, alongside a highly efficient non-DL method named MARS. The DL methods were trained on a common dataset of 3D cellular confocal microscopy images. Their segmentation accuracies were also tested in the presence of different image artefacts. A new method for segmentation quality evaluation was adopted which isolates segmentation errors due to under/over segmentation. This is complemented with new visualisation strategies that make interactive exploration of segmentation quality possible. Our analysis shows that the DL pipelines have very different levels of accuracy. Two of them show high performance, and offer clear advantages in terms of adaptability to new data.

show abstract

Section: Test Datasetmentioning

confidence: 99%

Benchmarking of deep learning algorithms for 3D instance segmentation of confocal image datasets

Kar

Petit

Refahi

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…In plants, spatio-temporal gene expression patterns are usually established using traditional in situ hybridization or by confocal microscopy of promoter fusions to fluorescent reporters. Confocal microscopy has the advantage that it can be used to reconstruct 3D structures by combining several z-stack images (Vijayan et al 2021; Hernandez-Lagana et al 2021; Wolny et al 2020; Bravo González-Blas et al 2020; Refahi et al 2021). In addition, combined with live image microscopy the temporal dynamics of gene expression and morphology development can be reconstructed (Refahi et al 2021; Valuchova et al 2020).…”

Section: Introductionmentioning

confidence: 99%

“…In addition, combined with live image microscopy the temporal dynamics of gene expression and morphology development can be reconstructed (Refahi et al 2021; Valuchova et al 2020). In this way, Refahi et al (2021) combined the information on spatio-temporal expression patterns of 28 regulatory genes into 3D reconstructed Arabidopsis flower meristems, ranging from initiation to stage 4 of flower development. These methods are limited by the low number of genes profiled per experiment, Therefore, tools to integrate scRNA-seq with expression data of defined, limited sets of 3D reference gene expression patterns need to be developed for spatial reconstruction single cell transcriptomes in plants.…”

Section: Introductionmentioning

confidence: 99%

“…Here, we adapted novoSpaRc (Nitzan et al 2019), a methodology for spatial reconstruction of single cell RNA-seq data, to generate a 3D single-cell transcriptome atlas of a floral meristem by integrating single nuclei RNA-seq and a 3D reconstructed flower meristem (Refahi et al 2021). NovoSpaRc reconstruction aims to explicitly preserve the transcriptome similarity among closely located scRNA-seq cells in the spatial map, while maximizing the transcriptome similarity between the scRNA-seq cells and the cells of the spatial map to which they are assigned.…”

Section: Introductionmentioning

confidence: 99%

“…In such a way, novoSpaRc performance is less affected by the number of reference genes than other methods, and, in theory, it can also be used without any reference gene (Nitzan et al 2019). However, novoSpaRc was developed to make use of spatial 2D continuous reference gene expression maps, while the 3D expression spatial map of floral meristem generated by Refahi et al 2021 is binary. We adapted the methodology for reconstructing single-cell transcriptomes in 3D making use of binary reference gene expression data.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A 3D gene expression atlas of the floral meristem based on spatial reconstruction of single nucleus RNA sequencing data

Neumann

Smaczniak

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Identity and functions of plant cells are influenced by their precise cellular location within the plant body. Cellular heterogeneity in growth and differentiation trajectories results in organ patterning. Therefore, assessing this heterogeneity at molecular scale is a major question in developmental biology. Single-cell transcriptomics (scRNA-seq) allows to characterize and quantify gene expression heterogeneity in developing organs at unprecedented resolution. However, the original physical location of the cell is lost during the scRNA-seq procedure. To recover the original location of cells is essential to link gene activity with cellular function and morphology. Here, we reconstruct genome-wide gene expression patterns of individual cells in a floral meristem by combining single-nuclei RNA-seq with 3D spatial reconstruction. By this, gene expression differences among meristematic domains giving rise to different tissue and organ types can be determined. As a proof of principle, the data are used to trace the initiation of vascular identity within the floral meristem. Our work demonstrates the power of spatially reconstructed single cell transcriptome atlases to understand plant morphogenesis. The floral meristem 3D gene expression atlas can be accessed at http://threed-flower-meristem.herokuapp.com

show abstract

Sepals

Roeder

2023

Encyclopedia of Life Sciences

View full text Add to dashboard Cite

Sepals are the outermost organs of a flower. They are sterile and generally green leaf‐like organs that surround and protect the developing reproductive structures inside the bud before the flower blooms. The sepals are the first organs initiated from the floral meristem and quickly grow to cover the floral meristem and initiating organ primordia. The sepals and petals in the flower interact with the environment by both attracting pollinators as well as defending against predators and abiotic factors such as the weather. Sepal organ identity is specified by the A function in the ABC model. Arabidopsis sepals are a model system for investigating morphogenesis (development of organ size and shape) and patterning (differentiation of specialised cell types such as giant cells). Current and future research on the development of sepals is leading to a better understanding of floral organ formation as well as underlying principles of cell division and growth. Key Concepts Sepals are the outermost floral organs, which are typically green leaf‐like organs that enclose and protect the developing flower bud before blooming. Sepals are thought to have evolved from leaves. Sepal organ identity is specified by A and E function in the ABC model of floral organ identity, and A function is specified by AP1/CAL/FUL clade MADS domain transcription factors, AP2 domain family transcription factors and microRNAs. The sepal is a commonly studied in Arabidopsis to discover fundamental principles of morphogenesis, such as how plant organs grow to reach the right size and shape. Arabidopsis sepals are the first floral organs to initiate on the flanks of the floral meristem where they quickly grow to cover and protect the developing reproductive organs, keep the bud closed until blooming and senesce and abscise after the fruit is fertilised. A scattered pattern of giant cells forms in between smaller cells in the outer epidermis of Arabidopsis sepals. The Arabidopsis thaliana MERISTEM LAYER1 (ATML1) transcription factor concentration fluctuates in epidermal cells of the developing sepal; if ATML1 reaches a high concentration during G2 phase of the cell cycle, the cell becomes giant. Giant cells become enlarged through endoreplication (also known as endoreduplication), during which a cell replicates its DNA and grows without division, becoming highly enlarged and polyploid. Sepal size and shape are highly reproducible; synchronous timing of both the initiation of sepal primordia and their maturation and cessation of growth are required for all four sepals in the Arabidopsis flower to maintain the same size to enclose the flower bud. The growth and division of sepal cells are highly variable, and gene expression is stochastic; however, sepals maintain robust size and shape through spatiotemporal averaging of growth variability and suppression of gene expression noise by the Paf1C complex.

show abstract

A multiscale analysis of early flower development in Arabidopsis provides an integrated view of molecular regulation and growth control

Cited by 44 publications

References 133 publications

Benchmarking of deep learning algorithms for 3D instance segmentation of confocal image datasets

Benchmarking of deep learning algorithms for 3D instance segmentation of confocal image datasets

A 3D gene expression atlas of the floral meristem based on spatial reconstruction of single nucleus RNA sequencing data

Sepals

Contact Info

Product

Resources

About