Kaptas, F. | Search bioRxiv

bioRxiv2025

A Morpho-Proteomic Atlas of Mitosis at Sub-Minute Resolution

Precise spatiotemporal protein organization is critical for fundamental biological processes including cell division1,2. Indeed, aberrant mitosis and mitotic factors are involved in diverse diseases, including various cancers3,4, Alzheimers disease5, and rare diseases6. During mitosis, complex spatial rearrangements and regulation ensure the accurate separation of replicated sister chromatids to produce genetically identical daughter cells7-9. Previous studies employed high-throughput methodologies to follow specific proteins during mitosis10-15. Still a temporally refined systems-level approach capable of monitoring morphological and proteomic changes throughout mitosis has been lacking. Here, we achieved unprecedented resolution by phenotypically decomposing mitosis into 40 subsections of a regression plane for proteomic analysis using deep learning and regression techniques. Our deep visual proteomics (DVP) workflow16, revealed rapid, dynamic proteomic changes throughout mitosis. We quantified 4,350 proteins with high confidence, demonstrating that 147 show significant dynamic abundance changes during mitotic progression. Clustering revealed coordinated patterns of protein regulation, while network analysis uncovered tight regulation of core cell cycle proteins and a link between cell cycle and cancer-linked mutations. Immunofluorescence validated abundance changes and linked previously uncharacterised proteins, like C19orf53, to mitosis. To facilitate data navigation, we developed Mito-Omix, a user-friendly online platform that integrates intricate morphological and molecular data. Our morphological and proteomic dataset spans mitosis at high resolution, providing a rich resource for understanding healthy and aberrant cell division.

cell biology↗

bioRxiv2022

Show me your neighbour and I tell what you are: fisheye transformation for deep learning-based single-cell phenotyping

Recently we have concluded that image-based features derived from the microenvironment have an enormous impact on successfully determining the class of an object1. Here we demonstrate that deep learning-based phenotypic analysis of cells with a properly chosen microenvironment-size provides results comparable to our earlier neighbourhood-based methods that utilise hand-crafted image features. We hypothesised that treating cells with equal weight, regardless of their position within the cellular microenvironment, is suboptimal, and direct neighbours have a larger impact on the phenotype of the cell-of-interest than cells in its larger proximity. Hence we present a novel approach that (1) considers the fully featured view of the cell-of-interest, (2) includes the neighbourhood and (3) gives lesser weight to cells that are far from the cell. To achieve this, we present a transformation similar to those characteristic for fisheye cameras. Such a transformation satisfies all the above defined criteria, with a fast rate of transform for any images. Using the proposed transformation with proper settings we could significantly increase the accuracy of single-cell phenotyping, both in case of cell culture and tissue-based microscopy images. The range of potential applications of the proposed method goes beyond microscopy, as we present improved results on the iWildCam 2020 dataset containing images of wild animals.

bioinformatics↗