Biology subjects

Traas, J.

Publications and source records attributed to Traas, J..

5 recordsLinked to original sources

bioRxiv2021

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

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

plant biology↗

bioRxiv2021

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

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 performances 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 3 dimensional (3D) cell 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 artifacts. A specific method for segmentation quality evaluation was adopted which isolates segmentation errors due to under/over segmentation. This is complemented with a 3D visualization strategy for interactive exploration of segmentation quality. Our analysis shows that the DL pipelines have different levels of accuracy. Two of them, which are end to end 3D and were originally designed for cell boundary detection, show high performance, and offer clear advantages in terms of adaptability to new data. Author summaryIn recent years a number of deep learning (DL) algorithms based on computational neural networks have been developed which claim to achieve high accuracy and automatic segmentation of 3D microscopy images. Although these algorithms have received considerable attention in the literature, it is difficult to evaluate their relative performances, while it remains unclear whether they really perform better than other, more classical segmentation methods. To clarify these issues, we performed a detailed, quantitative analysis of a number of representative DL pipelines for cell instance segmentation from 3D confocal microscopy image datasets. We developed a protocol for benchmarking the performances of such DL based segmentation pipelines using common training and test datasets, evaluation metrics and visualizations. Using this protocol, we evaluated and compared four different DL pipelines to identify their strengths and limitations. A high performance non-DL method was also included in the evaluation. We show that DL pipelines may show significant differences in their performances depending on their model architecture and pipeline components but overall show excellent adaptability to unseen data. We also show that our benchmarking protocol could be extended to a variety of segmentation pipelines and datasets.

bioinformatics↗

bioRxiv2020

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

The link between gene regulation and morphogenesis of multicellular organisms is a fundamental problem in biology. We address this question in the floral meristem of Arabidopsis, which generates new tissues and organs through complex changes in growth patterns. Starting from high-resolution time-lapse images, we generated a comprehensive 4-D atlas of early flower development including cell lineage, cellular growth rates and the expression patterns of 28 regulatory genes. This information was introduced in MorphoNet, a web-based open-access platform. The application of mechanistic computational models indicated that the molecular network based on the literature only explained a minority of the expression patterns. This was substantially improved by adding single regulatory hypotheses for individual genes. We next used the integrated information to correlate growth with the combinatorial expression of multiple genes. This led us to propose a set of hypotheses for the action of individual genes in morphogenesis, not visible by simply correlating gene expression and growth. This identified the central transcription factor LEAFY as a potential regulator of heterogeneous growth, which was supported by quantifying growth patterns in a leafy mutant. By providing an integrated, multiscale view of flower development, this atlas should represent a fundamental step towards mechanistic multiscale-scale models of flower development.

developmental biology↗

bioRxiv2019

Xyloglucan homeostasis and microtubule dynamics synergistically maintain meristem geometry and robustness of phyllotaxis in Arabidopsis

The shoot apical meristem (SAM) gives rise to all aerial organs of the plant. The cell walls are supposed to play a central role in this process, translating molecular regulation into dynamic changes of growth rates and directions, although their precise role in morphogenesis during organ formation remains not well understood. Here we investigate the role of xyloglucans (XyGs), which form a major, yet functionally poorly characterized, wall component in the SAM. Using immunolabeling, biochemical analysis, genetic approaches, micro-indentation, laser ablations and live imaging, we show that XyGs are important for meristem shape and phyllotaxis, although no difference in cell wall stiffness could be observed when XyGs are perturbed. Mutations in enzymes required for XyG synthesis also affect other cell wall components such as cellulose content and the pectin methylation status. Interestingly, we show that the control of cortical microtubules dynamics by the severing enzyme KATANIN becomes vital when XyGs are perturbed or absent. This suggests an active role of the cytoskeleton in compensating for altered wall composition.

plant biology↗

bioRxiv2019

A microtubule-mediated mechanical feedback controls leaf blade development in three dimensions

Many plant species have thin leaf blades, which is an important adaptation that optimizes the exchanges with the environment. Here, we provide evidence that their three-dimensional geometry is governed by microtubule alignment along mechanical stress patterns in internal walls. Depending on the primary shape of the primordium, this process has the potential to amplify an initial degree of flatness, or promote the formation of nearly axisymmetric, mostly elongating organs, such as stems and roots. This mechanism may explain leaf evolution from branches, which is alternative to Zimmermanns influential, but widely questioned, telome theory.\n\nOne Sentence SummaryMechanical feedback controls leaf development in three dimensions

developmental biology↗