# Reconstruction-Aware Cryo-EM Particle Picking

arXiv: https://arxiv.org/abs/2608.28838 · HTML: https://mlsanity.com/paper/2608.28838/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 71 references that are in the map.
- First posted on arXiv 28 Aug 2026.

## Builds on (most cited first, 10 of 10)

- [Positive-unlabeled convolutional neural networks for particle picking in cryo-electron micrographs](https://mlsanity.com/paper/1803.08207/) (2018)
- [DeepPicker: a Deep Learning Approach for Fully Automated Particle Picking in Cryo-EM](https://mlsanity.com/paper/1605.01838/) (2016)
- [A deep convolutional neural network approach to single-particle recognition in cryo-electron microscopy](https://mlsanity.com/paper/1605.05543/) (2016)
- [APPLE Picker: Automatic Particle Picking, a Low-Effort Cryo-EM Framework](https://mlsanity.com/paper/1802.00469/) (2018)
- [KLT Picker: Particle Picking Using Data-Driven Optimal Templates](https://mlsanity.com/paper/1912.06500/) (2019)
- [CryoMAE: Few-Shot Cryo-EM Particle Picking with Masked Autoencoders](https://mlsanity.com/paper/2404.10178/) (2024)
- [Unsupervised particle sorting for high-resolution single-particle cryo-EM](https://mlsanity.com/paper/1910.10051/) (2019)
- [ASOCEM: Automatic Segmentation Of Contaminations in cryo-EM](https://mlsanity.com/paper/2201.06978/) (2022)
- [Unsupervised particle sorting for cryo-EM using probabilistic PCA](https://mlsanity.com/paper/2210.12811/) (2022)
- [CRISP: A Framework for Cryo-EM Image Segmentation and Processing with Conditional Random Field](https://mlsanity.com/paper/2502.08287/) (2025)

## Abstract (arXiv)

Cryo-electron microscopy (cryo-EM) determines the structures of proteins and macromolecular assemblies at near-atomic resolution, and the final 3D reconstruction depends on extracting a clean particle stack from noisy micrographs. This extraction decomposes into three sub-tasks, namely particle picking, contamination removal, and 2D class selection. Each of them, however, is trained and evaluated in isolation, and none is optimized for the reconstruction. We instead integrate the three sub-tasks into a single pipeline posed against downstream reconstruction quality. We instantiate the pipeline with a state-of-the-art component for each sub-task, CryoTransformer picking permissively, MicrographCleaner masking contamination, and CryoSift selecting 2D classes by a continuous quality score, and close the loop with a fine-tuning step that returns the surviving particles to the picker. The pipeline achieves a better 3D resolution than every picker we compare. We also show that the best 2D F1 is not the best resolution, so particle selection is better treated as one reconstruction-aware pipeline judged by the map it delivers.