Unsupervised SE(3) Disentanglement for in situ Macromolecular Morphology Identification from Cryo-Electron Tomography

2026-01-04

arXiv.org

AI Summary

This work addresses the challenge of unsupervised in situ identification of macromolecular conformations from cryo-electron tomography (cryo-ET) data, which requires simultaneous estimation of structural templates and their SE(3) transformations. Existing approaches often fail to capture rare conformations and rely heavily on manual parameter tuning. To overcome these limitations, the authors propose a decoupled deep representation learning framework that disentangles SE(3) transformations from structural content in a latent space, augmented by a multi-choice learning mechanism to handle high-noise conditions. This method achieves, for the first time, unsupervised disentanglement of conformation and transformation directly from cryo-ET data, substantially enhancing the discovery of rare macromolecular states. It outperforms current methods on both simulated and real datasets, successfully revealing previously unknown macromolecular conformations.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Computer Vision: Representation Learning for Vision
• Reasoning under Uncertainty: Uncertainty Representations

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Security and Privacy: Data transparency and provenance
• Web Mining and Content Analysis: Bridging structured and unstructured data

Abstract

Cryo-electron tomography (cryo-ET) provides direct 3D visualization of macromolecules inside the cell, enabling analysis of their in situ morphology. This morphology can be regarded as an SE(3)-invariant, denoised volumetric representation of subvolumes extracted from tomograms. Inferring morphology is therefore an inverse problem of estimating both a template morphology and its SE(3) transformation. Existing expectation-maximization based solution to this problem often misses rare but important morphologies and requires extensive manual hyperparameter tuning. Addressing this issue, we present a disentangled deep representation learning framework that separates SE(3) transformations from morphological content in the representation space. The framework includes a novel multi-choice learning module that enables this disentanglement for highly noisy cryo-ET data, and the learned morphological content is used to generate template morphologies. Experiments on simulated and real cryo-ET datasets demonstrate clear improvements over prior methods, including the discovery of previously unidentified macromolecular morphologies.

Problem

• cryo-electron tomography
• macromolecular morphology
• SE(3) disentanglement
• in situ structure
• unsupervised learning

Innovation

• SE(3) disentanglement
• cryo-electron tomography
• unsupervised representation learning
• morphology identification
• multi-choice learning