Microscopy deep learning predicts virus infections and reveals mechanics of lytic-infected cells

• DOI: 10.1016/j.isci.2021.102543
• Publication Date: 2021-05-15T15:47:04Z
• Authors: Vardan Andriasyan; Artur Yakimovich; Anthony Petkidis; Fanny Georgi; Robert Witte; Daniel Puntener; Urs F. Greber
• Keywords: Lytic cycle

Imaging across scales reveals disease mechanisms in organisms, tissues, and cells. Yet, particular infection phenotypes, such as virus-induced cell lysis, have remained difficult to study. Here, we developed imaging modalities deep learning procedures identify herpesvirus adenovirus (AdV) infected cells without virus-specific stainings. Fluorescence microscopy of vital DNA-dyes live-cell revealed learnable nuclear patterns transferable related viruses the same family. Deep predicted two major AdV outcomes, non-lytic (nonspreading) lytic (spreading) infections, up about 20 hr prior lysis. Using these predictive algorithms, nuclei had levels green fluorescent protein (GFP)-tagged virion proteins but enriched faster, collapsed more extensively upon laser-rupture than nuclei, revealing impaired mechanical properties nuclei. Our algorithms may be used infer phenotypes emerging viruses, enhance single biology, facilitate differential diagnosis infections.