Single-particle reconstruction in cryo-EM based on three-dimensional weighted nuclear norm minimization
•Firstly, to overcome the over-smooth and the staircase effects of the TV-based reconstruction models, we apply the low rank-based model to better suppress the noise in the SPR task.•Secondly, treating different features the same is not conducive to the use of features, resulting in unsatisfactory r...
Saved in:
| Published in: | Pattern recognition Vol. 143; p. 109736 |
|---|---|
| Main Authors: | , , , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
Elsevier Ltd
01.11.2023
|
| Subjects: | |
| ISSN: | 0031-3203 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | •Firstly, to overcome the over-smooth and the staircase effects of the TV-based reconstruction models, we apply the low rank-based model to better suppress the noise in the SPR task.•Secondly, treating different features the same is not conducive to the use of features, resulting in unsatisfactory reconstruction results. We use the weighted nuclear norm to assign different weights to different features so that the model can make better use of features and reconstruct results with better structure and details.•Thirdly, as the SPR task has a three-dimensional structure, we extend the WNNM model to three-dimension and give the theoretical analysis of the three-dimensional weighted nuclear norm minimization (3DWNNM).•Fourthly, we design an efficient and theoretically guaranteed algorithm for the proposed model, which can solve the model accurately. Experiments on SPR in cryo-EM show that the proposed method can provide more outstanding reconstruction results than several state-of-the-art methods.
Single-particle reconstruction (SPR) in cryogenic electron microscopy (cryo-EM) aims at aligning and averaging two-dimensional micrographs to reconstruct a three-dimensional particle.
How to reconstruct micrographs from heavy noise is a crucial point for achieving better micrograph quality, and thus many methods focus on noise removal. However, new problems such as over-smoothing often occur in their results due to failure in handling heavy noise well. This paper proposes a three-dimensional weighted nuclear norm minimization (3DWNNM) model for SPR in the cryo-EM task to address these issues. Specifically, we design a minimization solver based on the forward-backward splitting algorithm to tackle our model efficiently. Under certain conditions, this solution has an energy-decaying feature and performs exceptionally well in reconstruction. Numerical experiments fully demonstrate the effectiveness and the robustness of the proposed method. |
|---|---|
| ISSN: | 0031-3203 |
| DOI: | 10.1016/j.patcog.2023.109736 |