Influenza viruses pose severe public health threats globally. Influenza viruses are extensively pleomorphic, in shape, size, and organization of viral proteins. Analysis of influenza morphology and ultrastructure can help elucidate viral structure-function relationships and aid in therapeutics and vaccine development. While cryo-electron tomography (cryoET) can depict the 3D organization of pleomorphic influenza, the low signal-to-noise ratio inherent to cryoET and viral heterogeneity have precluded detailed characterization of influenza viruses. In this report, we leveraged convolutional neural networks and cryoET to characterize the morphological architecture of the A/Puerto Rico/8/34 (H1N1) influenza strain. Our pipeline improved the throughput of cryoET analysis and accurately identified viral components within tomograms. Using this approach, we successfully characterized influenza morphology, glycoprotein density, and conducted subtomogram averaging of influenza glycoproteins. Application of this processing pipeline can aid in the structural characterization of not only influenza viruses, but other pleomorphic viruses and infected cells.

流感病毒对全球公共卫生构成严重威胁。流感病毒在形状、大小和病毒蛋白组织方面具有广泛的多形性。流感形态学和超微结构分析有助于阐明病毒结构-功能关系，并有助于治疗和疫苗开发。虽然低温电子断层扫描 (cryoET) 可以描绘多形性流感的 3D 组织，但 cryoET 固有的低信噪比和病毒异质性阻碍了对流感病毒的详细表征。在本报告中，我们利用卷积神经网络和 cryoET 来表征 A/Puerto Rico/8/34 (H1N1) 流感病毒株的形态结构。我们的流程提高了 cryoET 分析的吞吐量，并准确识别了断层图中的病毒成分。使用这种方法，我们成功地表征了流感形态、糖蛋白密度，并对流感糖蛋白进行了子断层图平均。这种处理管道的应用不仅可以帮助分析流感病毒的结构特征，还可以帮助分析其他多形性病毒和受感染细胞的结构特征。