A deep learning-based toolkit for 3D nuclei segmentation and quantitative analysis in cellular and tissue context.

We present a new set of computational tools that enable accurate and widely applicable 3D segmentation of nuclei in various 3D digital organs. We have developed an approach for ground truth generation and iterative training of 3D nuclear segmentation models, which we applied to popular CellPose, PlantSeg and StarDist algorithms. We provide two high-quality models trained on plant nuclei that enable 3D segmentation of nuclei in datasets obtained from fixed or live samples, acquired from different plant and animal tissues, and stained with various nuclear stains or fluorescent protein-based nuclear reporters. We also share a diverse high-quality training dataset of about 10,000 nuclei. Furthermore, we advanced the MorphoGraphX analysis and visualization software by, among other things, providing a method for linking 3D segmented nuclei to their surrounding cells in 3D digital organs. We found that the nuclear-to-cell volume ratio varies between different ovule tissues and during the development of a tissue. Finally, we extended the PlantSeg 3D segmentation pipeline with a proofreading tool that uses 3D segmented nuclei as seeds to correct cell segmentation errors in difficult-to-segment tissues.

Visualizing single-cell data with the neighbor embedding spectrum.

The two-dimensional embedding methods t-SNE and UMAP are ubiquitously used for visualizing single-cell data. Recent theoretical research in machine learning has shown that, despite their very different formulation and implementation, t-SNE and UMAP are closely connected, and a single parameter suffices to interpolate between them. This leads to a whole spectrum of visualization methods that focus on different aspects of the data. Along the spectrum, this focus changes from representing local structures to representing continuous ones. In single-cell context, this leads to a trade-off between highlighting rare cell types or continuous variation, such as developmental trajectories. Visualizing the entire spectrum as an animation can provide a more nuanced understanding of the high-dimensional dataset than individual visualizations with either t-SNE or UMAP.

Scalable Optical Nose Realized with a Chemiresistively Modulated Light-Emitter Array.

Biological olfaction relies on a large number of receptors that function as sensors to detect gaseous molecules. It is challenging to realize artificial olfactory systems that contain similarly large numbers of sensory materials. It is shown that combinatorial materials processing with vapor deposition can be used to fabricate large arrays of distinct chemiresistive sensing materials. By combining these with light-emitting diodes, an array of chemiresistively-modulated light-emitting diodes, or ChemLEDs, that permit a simultaneous optical read-out in response to an analyte is obtained. The optical nose uses a common voltage source and ground for all sensing elements and thus eliminates the need for complex wiring of individual sensors. This optical nose contains one hundred ChemLEDs and generates unique light patterns in response to gases and their mixtures. Optical pattern recognition methods enable the quantitative prediction of the corresponding concentrations and compositions, thereby paving the way for massively parallel artificial olfactory systems. ChemLEDs open the possibility to explore demanding gas sensing applications, including in environmental, food quality monitoring, and potentially diagnostic settings.