Particle Identification (Semantic Candidates → Particle Centers)¶

The Particle Identification task in CryoSiam is designed to detect candidate macromolecular complexes and convert them into particle coordinates suitable for downstream workflows (csv and star files).

It works in two steps:

Semantic prediction of candidate classes (multi-class semantic model)
Conversion to particle centers (extracting coordinates from high-confidence regions)

Workflow:

Denoising → Lamella Prediction → Semantic (particle candidates) → Particle centers (.star / .csv)

Recommended: Always use a lamella mask to remove false positives outside the lamella, since models are trained only on lamella regions.

Example Results¶

Semantic particle prediction (centers):

Trained Models¶

You can download the trained lamella model from here: CryoSiam lamella model (v1.0), and the semantic segmentation model from here: CryoSiam particle identification model (v1.0). You can also train your own model and then perform prediction with that model. Review the Semantic training for explanation of the model training procedure.

Commands¶

Semantic Prediction for Particle Candidates¶

cryosiam semantic_predict --config_file=configs/semantic_particle.yaml

Convert Semantic Predictions to Particle Centers¶

cryosiam semantic_to_centers --config_file=configs/semantic_particle.yaml

Visualization For visualization of the results, CryoSiam-Vis can be used as described here.

Example Configuration (`configs/semantic_particle.yaml`)¶

data_folder: '/scratch/stojanov/dataset1/predictions/denoised'
mask_folder: '/scratch/stojanov/dataset1/predictions/lamella'
log_dir: '/scratch/stojanov/dataset1/'
prediction_folder: '/scratch/stojanov/dataset1/predictions/semantic_particle'

trained_model: '/scratch/stojanov/trained_models/cryosiam_semantic_myco_candidates.ckpt'
file_extension: '.mrc'

test_files: null

parameters:
  gpu_devices: 1
  data:
    patch_size: [ 128, 128, 128 ]
    min: 0
    max: 1
    mean: 0
    std: 1
  network:
    in_channels: 1
    spatial_dims: 3
    out_channels: 14
    threshold: 0.7
    min_voxels: 10

hyper_parameters:
  batch_size: 2

Config Reference¶

Top-level keys¶

Key	Type	Must change	Description
`data_folder`	`str`	✅	Folder with denoised tomograms.
`mask_folder`	`str`	✅	Folder with lamella masks to restrict predictions.
`log_dir`	`str`	❌	Where logs and debug files are written.
`prediction_folder`	`str`	✅	Output folder for semantic predictions + particle centers.
`trained_model`	`str`	✅	Path to `.ckpt` particle‑candidate model.
`file_extension`	`str`	❌	`.mrc` or `.rec`.
`test_files`	`list[str]` or `null`	❌	Which tomograms to run.

`parameters`¶

Key	Type	Must change	Description
`gpu_devices`	`int/list[int]`	❌	GPU to use.
`data.patch_size`	`list[int]`	❌	Patch size for inference.
`data.min` / `data.max`	`float`	❌	Intensity clipping.
`data.mean` / `data.std`	`float`	❌	Normalization stats.
`network.in_channels`	`int`	❌	Number of input channels (always 1).
`network.spatial_dims`	`int`	❌	3 for 3D tomograms.
`network.out_channels`	`int`	✅	Number of semantic classes (14).
`network.threshold`	`float`	❌	Probability threshold for detection.
`network.min_voxels`	`int`	❌	Minimum number of voxels to consider as valid prediction. Default is 100

`hyper_parameters`¶

Key	Type	Must change	Description
`batch_size`	`int`	❌	Number of patches processed per batch.