Disentangling semantic features of macromolecules in Cryo-Electron Tomography
This work addresses the problem of systematic macromolecule analysis in Cryo-ET for researchers in structural biology, but it appears incremental as it builds on existing disentanglement methods for a specific domain.
The paper tackled the challenge of recognizing and recovering macromolecular structures in Cryo-Electron Tomography by proposing a 3D Spatial Variational Autoencoder that disentangles structure, orientation, and shift features, demonstrating efficacy through experiments on synthesized and real datasets.
Cryo-electron tomography (Cryo-ET) is a 3D imaging technique that enables the systemic study of shape, abundance, and distribution of macromolecular structures in single cells in near-atomic resolution. However, the systematic and efficient $\textit{de novo}$ recognition and recovery of macromolecular structures captured by Cryo-ET are very challenging due to the structural complexity and imaging limits. Even macromolecules with identical structures have various appearances due to different orientations and imaging limits, such as noise and the missing wedge effect. Explicitly disentangling the semantic features of macromolecules is crucial for performing several downstream analyses on the macromolecules. This paper has addressed the problem by proposing a 3D Spatial Variational Autoencoder that explicitly disentangle the structure, orientation, and shift of macromolecules. Extensive experiments on both synthesized and real cryo-ET datasets and cross-domain evaluations demonstrate the efficacy of our method.