CryoSPIN: Improving Ab-Initio Cryo-EM Reconstruction with Semi-Amortized Pose Inference

TL;DR

CryoSPIN enhances cryo-EM 3D reconstruction by combining initial amortized pose inference with subsequent local pose refinement, leading to faster and more accurate results than existing methods.

Contribution

This paper introduces cryoSPIN, a semi-amortized approach that improves pose inference in cryo-EM by integrating amortized inference with auto-decoding refinement.

Findings

Handles multi-modal pose distributions effectively.

Faster convergence of pose estimation.

Outperforms cryoAI in speed and quality.

Abstract

Cryo-EM is an increasingly popular method for determining the atomic resolution 3D structure of macromolecular complexes (eg, proteins) from noisy 2D images captured by an electron microscope. The computational task is to reconstruct the 3D density of the particle, along with 3D pose of the particle in each 2D image, for which the posterior pose distribution is highly multi-modal. Recent developments in cryo-EM have focused on deep learning for which amortized inference has been used to predict pose. Here, we address key problems with this approach, and propose a new semi-amortized method, cryoSPIN, in which reconstruction begins with amortized inference and then switches to a form of auto-decoding to refine poses locally using stochastic gradient descent. Through evaluation on synthetic datasets, we demonstrate that cryoSPIN is able to handle multi-modal pose distributions during the amortized inference stage, while the later, more flexible stage of direct pose optimization yields faster and more accurate convergence of poses compared to baselines. On experimental data, we show that cryoSPIN outperforms the state-of-the-art cryoAI in speed and reconstruction quality.