Bekbayev, S. | Search bioRxiv

bioRxiv2026

Super-Resolution Processing Enables In Situ Subtomogram Averaging Beyond the Nyquist Limit

In situ subtomogram averaging (STA) for high-resolution structure determination requires fine pixel sampling, forcing a trade-off between achievable resolution and the field of view (FOV) recorded in each tomogram. Electron-event representation (EER) data can be rendered at half the physical pixel size, but whether super-resolution processing can recover useful information beyond the physical Nyquist limit from noisy in situ data remains unclear. Here, we show that super-resolution processing of EER data recovers biologically interpretable information beyond a physical Nyquist limit of 4.78 A in two in situ datasets. STA of unidentified virus particles in a Drosophila melanogaster larva yielded a 3.42 A inner capsid reconstruction, enabling de novo model building and identification of the particles as Drosophila A virus. In situ STA of ribosomes in Saccharomyces cerevisiae reached 3.18 A, with local resolution extending to 2.78 A. EER acquisition therefore provides greater flexibility in choosing the acquisition magnification: existing datasets can be re-rendered at the super-resolution pixel size when physical sampling becomes limiting, while lower-magnification acquisition can be considered when a larger FOV improves target capture or cellular context.

molecular biology↗

bioRxiv2023

CebraEM: A practical workflow to segmentcellular organelles in volume SEM datasetsusing a transferable CNN-based membraneprediction

Segmentation of large-volume datasets obtained by volume SEM techniques is a challenging task that generally requires a considerable amount of human effort. Despite recent advances in deep learning leading to the successful segmentation of cellular organelles in a variety of datasets, it is still challenging and time-consuming to produce the necessary data for training a convolutional neural network as well as to set up targeted post-processing pipelines to obtain a good quality full-volume semantic instance segmentation. We present CebraEM, a software package that uses a novel workflow for the segmentation of organelles in volume EM datasets, which helps to minimize the annotation time for the generation of training data. It relies on a generic CNN-based membrane prediction, followed by a well-established machine-learning pipeline that includes over-segmentation before random forest classification and graph multi-cut grouping. The workflow was tested for the segmentation of organelles on different datasets originating from various sample preparations and imaging modalities in volume SEM, in each case resulting in state-of-the-art semantic instance segmentations without additional post-processing. Importantly, by considerably simplifying the segmentation problem, CebraEM empowers single users with the ability to efficiently segment hundreds of gigabytes of data.

cell biology↗