arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.21915cs.CV
收藏

基于蛋白质图神经网络的异质冷冻电镜重建

Protein Graph Neural Networks for Heterogeneous Cryo-EM Reconstruction

KTH Royal Institute of Technology(皇家理工学院);Harvard University(哈佛大学);Department of Mathematics(数学系);School of Engineering and Applied Sciences(工程与应用科学学院)

Jonathan Krook, Axel Janson, Joakim Andén, Melanie Weber, Ozan Öktem

AI总结:

本文提出基于图神经网络的异质冷冻电镜重建方法,通过几何感知的归纳偏置提升骨架构象预测的准确性。

摘要:

我们提出了一种几何感知的方法,用于异质单颗粒冷冻电镜(cryo-EM)重建,以预测原子骨架构象。为了纳入蛋白质结构先验知识,我们将骨架表示为图,并使用图神经网络(GNN)自动解码器,将每张图像的潜在变量映射到模板构象的3D位移。目标结合了基于可微分cryo-EM前向模型的数据不一致项和几何正则化,并支持未知取向通过椭球支持提升(ESL)姿态估计。在从分子动力学轨迹衍生出的合成数据集上,所提出的GNN在准确度上优于具有相似规模的多层感知机(MLP),突显了几何感知归纳偏置的优势。

英文摘要:

We present a geometry-aware method for heterogeneous single-particle cryogenic electron microscopy (cryo-EM) reconstruction that predicts atomic backbone conformations. To incorporate protein-structure priors, we represent the backbone as a graph and use a graph neural network (GNN) autodecoder that maps per-image latent variables to 3D displacements of a template conformation. The objective combines a data-discrepancy term based on a differentiable cryo-EM forward model with geometric regularization, and it supports unknown orientations via ellipsoidal support lifting (ESL) pose estimation. On synthetic datasets derived from molecular dynamics trajectories, the proposed GNN achieves higher accuracy compared to a multilayer perceptron (MLP) of comparable size, highlighting the benefits of a geometry-informed inductive bias.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑