FakET: Simulating Cryo-Electron Tomograms with Neural Style Transfer

TL;DR

FakET uses neural style transfer to rapidly generate high-quality simulated cryo-electron microscopy data, significantly reducing time and memory requirements while maintaining performance for particle localization and classification tasks.

Contribution

The paper introduces FakET, a novel neural style transfer-based method for fast, scalable simulation of cryo-EM data that matches real data performance with much higher efficiency.

Findings

Data generation speed increased 750 times

Memory usage reduced 33 times

Maintains state-of-the-art localization and classification performance

Abstract

In cryo-electron microscopy, accurate particle localization and classification are imperative. Recent deep learning solutions, though successful, require extensive training data sets. The protracted generation time of physics-based models, often employed to produce these data sets, limits their broad applicability. We introduce FakET, a method based on Neural Style Transfer, capable of simulating the forward operator of any cryo transmission electron microscope. It can be used to adapt a synthetic training data set according to reference data producing high-quality simulated micrographs or tilt-series. To assess the quality of our generated data, we used it to train a state-of-the-art localization and classification architecture and compared its performance with a counterpart trained on benchmark data. Remarkably, our technique matches the performance, boosts data generation speed 750 times, uses 33 times less memory, and scales well to typical transmission electron microscope detector sizes. It leverages GPU acceleration and parallel processing. The source code is available at https://github.com/paloha/faket.