Reconstructing Heterogeneous Biomolecules via Hierarchical Gaussian Mixtures and Part Discovery

2025-06-06

AI Summary

This study addresses the challenge of 3D reconstruction of non-rigid biomacromolecules in cryo-electron microscopy (cryo-EM), where strong heterogeneity arises jointly from conformational flexibility and compositional incompleteness (e.g., missing subunits). We propose CryoSPIRE, a novel hierarchical Gaussian mixture model. Methodologically, it introduces data-driven component discovery as an inductive bias to jointly model conformational and compositional heterogeneity; integrates Gaussian splatting, hierarchical Gaussian mixture representation, and differentiable 3D reconstruction optimization. On the CryoBench benchmark, CryoSPIRE achieves new state-of-the-art performance. It successfully resolves biologically meaningful dynamic conformations and missing-component states in multiple real experimental datasets, significantly improving both reconstruction accuracy and interpretability for highly heterogeneous samples.

Technology Category

Cognitive Modeling & Cognitive Systems: Neural Spike CodingMachine Learning: Mixture of Experts (MoE)Computer Vision: 3D Computer Vision

Application Category

Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphsSecurity and Privacy: Data transparency and provenanceSemantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Cryo-EM is a transformational paradigm in molecular biology where computational methods are used to infer 3D molecular structure at atomic resolution from extremely noisy 2D electron microscope images. At the forefront of research is how to model the structure when the imaged particles exhibit non-rigid conformational flexibility and compositional variation where parts are sometimes missing. We introduce a novel 3D reconstruction framework with a hierarchical Gaussian mixture model, inspired in part by Gaussian Splatting for 4D scene reconstruction. In particular, the structure of the model is grounded in an initial process that infers a part-based segmentation of the particle, providing essential inductive bias in order to handle both conformational and compositional variability. The framework, called CryoSPIRE, is shown to reveal biologically meaningful structures on complex experimental datasets, and establishes a new state-of-the-art on CryoBench, a benchmark for cryo-EM heterogeneity methods.

Problem

• Reconstructing biomolecules with conformational flexibility and missing parts
• Modeling non-rigid structural variations in cryo-EM data
• Handling compositional variability in heterogeneous molecular structures

Innovation

• Hierarchical Gaussian mixture model
• Part-based segmentation inference
• CryoSPIRE framework