topaz
Pipeline for particle picking in cryo-electron microscopy images using convolutional neural networks trained from positive and unlabeled examples. Also featurin
What is topaz?
Pipeline for particle picking in cryo-electron microscopy images using convolutional neural networks trained from positive and unlabeled examples. Also featuring micrograph and tomogram denoising with DNNs.
Overview of topaz
What it is
Topaz is a pipeline for particle detection in cryo-electron microscopy (cryo-EM) images. It uses convolutional neural networks trained from positive and unlabeled examples. The software also includes methods for micrograph and tomogram denoising using deep denoising models.
Key features
- Particle picking in cryo-EM micrographs using convolutional neural networks
- Micrograph and tomogram denoising with deep neural networks
- Pretrained particle picking and denoising models included
- Integration with Relion via provided scripts
- Graphical user interface (GUI) for ease of use
- Supports 3D denoising and particle extraction in patches
- Can read long argument lists using @list_of_many_files.txt
Use cases
- Automated particle picking from cryo-EM micrographs
- Denoising of cryo-EM micrographs and tomograms to improve image quality
- Training custom particle detection models from user-provided positive and unlabeled examples
Getting started
Install Topaz using Anaconda (recommended):conda create -n topaz python=3.8source activate topazconda install topaz -c tbepler -c pytorchOr install via pip:pip3 install topaz-emFor Docker, build from the repository:docker build -t topaz https://github.com/tbepler/topaz.gitFor Singularity:singularity pull shub://nysbc/topazSee the repository README for detailed installation instructions and tutorials.
License and pricing
Topaz is licensed under GPL-3.0 and is open source. No pricing information is provided beyond the open-source license.
Based on the project's public README and repository metadata.