Saved in:
Bibliographic Details
Main Authors: Jiang, Runmin, Zhang, Genpei, Yang, Yuntian, Wu, Siqi, Wu, Minhao, Feng, Wanyue, Zhao, Yizhou, Xiao, Xi, Wang, Xiao, Wang, Tianyang, Li, Xingjian, Chen, Muyuan, Xu, Min
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.23444
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914074428702720
author Jiang, Runmin
Zhang, Genpei
Yang, Yuntian
Wu, Siqi
Wu, Minhao
Feng, Wanyue
Zhao, Yizhou
Xiao, Xi
Wang, Xiao
Wang, Tianyang
Li, Xingjian
Chen, Muyuan
Xu, Min
author_facet Jiang, Runmin
Zhang, Genpei
Yang, Yuntian
Wu, Siqi
Wu, Minhao
Feng, Wanyue
Zhao, Yizhou
Xiao, Xi
Wang, Xiao
Wang, Tianyang
Li, Xingjian
Chen, Muyuan
Xu, Min
contents Single-particle cryo-electron microscopy (cryo-EM) has become a cornerstone of structural biology, enabling near-atomic resolution analysis of macromolecules through advanced computational methods. However, the development of cryo-EM processing tools is constrained by the scarcity of high-quality annotated datasets. Synthetic data generation offers a promising alternative, but existing approaches lack thorough biophysical modeling of heterogeneity and fail to reproduce the complex noise observed in real imaging. To address these limitations, we present CryoCCD, a synthesis framework that unifies versatile biophysical modeling with the first conditional cycle-consistent diffusion model tailored for cryo-EM. The biophysical engine provides multi-functional generation capabilities to capture authentic biological organization, and the diffusion model is enhanced with cycle consistency and mask-guided contrastive learning to ensure realistic noise while preserving structural fidelity. Extensive experiments demonstrate that CryoCCD generates structurally faithful micrographs, enhances particle picking and pose estimation, as well as achieves superior performance over state-of-the-art baselines, while also generalizing effectively to held-out protein families.
format Preprint
id arxiv_https___arxiv_org_abs_2505_23444
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CryoCCD: Conditional Cycle-consistent Diffusion with Biophysical Modeling for Cryo-EM Synthesis
Jiang, Runmin
Zhang, Genpei
Yang, Yuntian
Wu, Siqi
Wu, Minhao
Feng, Wanyue
Zhao, Yizhou
Xiao, Xi
Wang, Xiao
Wang, Tianyang
Li, Xingjian
Chen, Muyuan
Xu, Min
Computer Vision and Pattern Recognition
Artificial Intelligence
Single-particle cryo-electron microscopy (cryo-EM) has become a cornerstone of structural biology, enabling near-atomic resolution analysis of macromolecules through advanced computational methods. However, the development of cryo-EM processing tools is constrained by the scarcity of high-quality annotated datasets. Synthetic data generation offers a promising alternative, but existing approaches lack thorough biophysical modeling of heterogeneity and fail to reproduce the complex noise observed in real imaging. To address these limitations, we present CryoCCD, a synthesis framework that unifies versatile biophysical modeling with the first conditional cycle-consistent diffusion model tailored for cryo-EM. The biophysical engine provides multi-functional generation capabilities to capture authentic biological organization, and the diffusion model is enhanced with cycle consistency and mask-guided contrastive learning to ensure realistic noise while preserving structural fidelity. Extensive experiments demonstrate that CryoCCD generates structurally faithful micrographs, enhances particle picking and pose estimation, as well as achieves superior performance over state-of-the-art baselines, while also generalizing effectively to held-out protein families.
title CryoCCD: Conditional Cycle-consistent Diffusion with Biophysical Modeling for Cryo-EM Synthesis
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2505.23444