Deposition
Octopi and TopCup particle picks for a self-configuring in situ cryoET structure-determination workflow
- Deposition ID:CZCDP-10358
Release Date: 2026-06-24
Last Modified: 2026-06-24

Photo Caption: Central slab of 4 representative tomograms with selected sets of annotations.
Deposition Overview
Particle picks produced by Octopi and TopCup, two deep-learning particle-detection models from the self-configuring in situ cryoET structure-determination workflow (Octopi + py2rely). Octopi is a self-configuring 3D U-Net whose architecture/hyperparameters are selected per dataset by Bayesian optimization; TopCup is an open re-implementation of the top-ranked 2024-2025 Kaggle CryoET Object Identification Challenge solution, used as a fixed-architecture baseline. Picks are deposited for: GroEL in in-situ E. coli; apo-ferritin, beta-amylase, beta-galactosidase, ribosome, thyroglobulin and virus-like particle in the Phantom/ML-Challenge datasets (both Octopi and TopCup); cytosolic and membrane-bound ribosomes in unroofed cells; and V-ATPase on isolated synaptic vesicles. All picks are model predictions validated by subtomogram averaging with py2rely (RELION-5).
Authors
Deposition Data
Annotations:6,372
Tomograms:0
Publications
Not Submitted
Related Databases
Not Submitted
Methods Summary
Annotations | Method Type | Method Details | Method Links |
|---|---|---|---|
| 2968 | Automated | Automated particle detection with Octopi, a self-configuring supervised 3D U-Net (built on MONAI primitives) whose architecture and training hyperparameters are selected per dataset via Bayesian optimization (Optuna, TPE sampler). Particle coordinates were extracted from the predicted per-class segmentation masks using a watershed-based peak detector, retaining candidates whose effective radius is 50-100% of the specified particle radius. Coordinates were validated by subtomogram averaging with py2rely (RELION-5). | Source Code:copick +4 more |
| 301 | Automated | Automated particle detection with Octopi, a self-configuring supervised 3D U-Net (built on MONAI primitives) whose architecture and training hyperparameters are selected per dataset via Bayesian optimization (Optuna, TPE sampler). Particle coordinates were extracted from the predicted per-class segmentation masks using a watershed-based peak detector, retaining candidates whose effective radius is 50-100% of the specified particle radius. Coordinates were validated by subtomogram averaging with py2rely (RELION-5). For this membrane-associated target, membrane-proximal particles were separated from distal ones by Euclidean distance to MemBrain-seg-segmented membranes, and each pick was assigned an initial orientation from the local membrane normal. | Source Code:copick +5 more |
| 2841 | Automated | Automated particle detection with TopCup, an open re-implementation of the top-ranked solution of the 2024-2025 Kaggle CryoET Object Identification Challenge: a 3D U-Net (MONAI FlexibleUNet with a ResNet/EfficientNet encoder) with a segmentation head at each decoder level (deep supervision), trained on binary-sphere targets with dense cross-entropy loss. Coordinates were validated by subtomogram averaging with py2rely (RELION-5). | Source Code:copick +4 more |
| 262 | Automated | MemBrain-seg membrane segmentation at the native tomogram resolution, cleaned by clipping to the sample boundaries. | Source Code:copick +1 more |
Deposited Data
Data Contents
Filter by:
6,372 of 6,372 Annotations
Annotation Name | Object Shape Type | Method Type | Deposited In |
|---|---|---|---|
169virus-like capsid Annotation ID: AN-181572 | OrientedPoint | Automated | |
168virus-like capsid Annotation ID: AN-181571 | OrientedPoint | Automated | |
167Thyroglobulin Annotation ID: AN-181570 | OrientedPoint | Automated | |
166Thyroglobulin Annotation ID: AN-181569 | OrientedPoint | Automated | |
165cytosolic ribosome Annotation ID: AN-181568 | OrientedPoint | Automated | |
164cytosolic ribosome Annotation ID: AN-181567 | OrientedPoint | Automated | |
163Beta-galactosidase Annotation ID: AN-181566 | OrientedPoint | Automated | |
162Beta-galactosidase Annotation ID: AN-181565 | OrientedPoint | Automated | |
161Beta-amylase Annotation ID: AN-181564 | OrientedPoint | Automated | |
160Beta-amylase Annotation ID: AN-181563 | OrientedPoint | Automated | |
159ferritin complex Annotation ID: AN-181562 | OrientedPoint | Automated | |
158ferritin complex Annotation ID: AN-181561 | OrientedPoint | Automated | |
169virus-like capsid Annotation ID: AN-181560 | OrientedPoint | Automated | |
168virus-like capsid Annotation ID: AN-181559 | OrientedPoint | Automated | |
167Thyroglobulin Annotation ID: AN-181558 | OrientedPoint | Automated | |
166Thyroglobulin Annotation ID: AN-181557 | OrientedPoint | Automated | |
165cytosolic ribosome Annotation ID: AN-181556 | OrientedPoint | Automated | |
164cytosolic ribosome Annotation ID: AN-181555 | OrientedPoint | Automated | |
163Beta-galactosidase Annotation ID: AN-181554 | OrientedPoint | Automated | |
162Beta-galactosidase Annotation ID: AN-181553 | OrientedPoint | Automated |
- 1
- 2
- 3
- 4
- 5
- 319