Rapid coarse-grained simulation of whole cryo-tomograms

Purnell, Carson; Heebner, Jessica; Swulius, Matthew T.

doi:10.1016/j.bpj.2022.11.916

Cited by 1 publication

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• CryoTomoSim (Purnell et al, 2023): Utilizing this simulation tool, we generated 20 new training patches and 10 for testing purposes. For the generation of these patches, we varied the simulated tomogram acquisition parameters (e.g.…”

Section: Our Integrated Datasetsmentioning

confidence: 99%

“…This expansion added 20 DeePiCt patches (Table 1 Round 4), where we extracted patches in regions where both MemBrain-seg's performance was limited and the available ground truth was well-annotated. Additionally, we generated 20 patches using synthetic tomogram generators CryoTomoSim (Purnell et al, 2023) and PolNet (Martinez-Sanchez, Jasnin, et al, 2023), respectively (Table 1 Round 5).…”

Section: Iterative Approach For Generating Initial Training Datamentioning

confidence: 99%

“…The segmentation of membranes in cryo-electron tomograms poses significant challenges, notably due to the low signal-to-noise ratio inherent to cryo-ET data. Various denoising methods have been employed, including classical Wiener-like deconvolution (Tegunov and Cramer, 2019), as well as innovative deep learning-based approaches (Buchholz et al, 2019; Purnell et al, 2023; Bepler et al, 2020; Maldonado et al, 2023). For example, Cryo-CARE builds a noise model by comparing even and odd frames of a tilt-series (Buchholz et al, 2019), and hence can denoise images without ever seeing a noise-free example.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

MemBrain v2: an end-to-end tool for the analysis of membranes in cryo-electron tomography

Lamm,

Zufferey,

Righetto

et al. 2024

Preprint

View full text Add to dashboard Cite

MemBrain v2 is a deep learning-enabled program aimed at the efficient analysis of membranes in cryo-electron tomography (cryo-ET). The final v2 release of MemBrain will comprise three main modules: 1) MemBrain-seg, which provides automated membrane segmentation, 2) MemBrain-pick, which provides automated picking of particles along segmented membranes, and 3) MemBrain-stats, which provides quantitative statistics of particle distributions and membrane morphometrics.This initial version of the manuscript is focused on the beta release of MemBrain-seg, which combines iterative training with diverse data and specialized Fourier-based data augmentations. These augmentations are specifically designed to enhance the tool’s adaptability to a variety of tomographic data and address common challenges in cryo-ET analysis. A key feature of MemBrain-seg is the implementation of the Surface-Dice loss function, which improves the network’s focus on membrane connectivity and allows for the effective incorporation of manual annotations from different sources. This function is beneficial in handling the variability inherent in membrane structures and annotations. Our ongoing collaboration with the cryo-ET community plays an important role in continually improving MemBrain v2 with a wide array of training data. This collaborative approach ensures that MemBrain v2 remains attuned to the field’s needs, enhancing its robustness and generalizability across different types of tomographic data.The current version of MemBrain-seg is available athttps://github.com/teamtomo/membrainseg, and the predecessor of MemBrain-pick (also called MemBrain v1) is deposited athttps://github.com/CellArchLab/MemBrain. This preprint will be updated concomitantly with the code until the three integrated modules of MemBrain v2 are complete.

show abstract

Section: Our Integrated Datasetsmentioning

confidence: 99%

Section: Iterative Approach For Generating Initial Training Datamentioning

confidence: 99%