Topaz (Bepler, et al)

Overview of the Topaz wrapper available through CryoSPARC.

Introduction

The Topaz wrapper in CryoSPARC incorporates deep learning models used in Topaz to automatically pick particles with a set of previously-picked particles or to denoise micrographs. The wrapper consists of four jobs:

• Topaz Train

• Topaz Cross Validation

• Topaz Extract

• Topaz Denoise

Use the first three jobs for particle picking. Use final job for micrograph denoising.

Topaz is a particle detection tool created by Tristan Bepler, Alex J. Noble and team:

Bepler, T., Morin, A., Rapp, M. et al. Positive-unlabeled convolutional neural networks for particle picking in cryo-electron micrographs. Nat Methods 16, 1153–1160 (2019) doi:10.1038/s41592-019-0575-8

Bepler, T., Noble, A.J., Berger, B. Topaz-Denoise: general deep denoising models for cryoEM. bioRxiv 838920 (2019) doi: <https://doi.org/10.1101/838920>

Topaz License

Structura Biotechnology Inc. and CryoSPARC do not license Topaz nor distribute Topaz binaries. Please ensure you have your own copy of Topaz licensed and installed under the terms of its GNU General Public License v3.0.

Installing Topaz

CryoSPARC requires a Topaz installation in a dedicated conda environment. The steps below are based on Topaz developers' instructions and were modified in December 2024 to force installation of Topaz v0.2.5 and compatible libraries. Newer versions of Topaz are available, but may not be compatible with the Topaz wrapper in CryoSPARC.

Create an conda Python environment

Do not use the conda Python installed with CryoSPARC. This installation is destroyed during CryoSPARC updates.

Important considerations for Master/Worker or Cluster installations:

• The path to the Anaconda installation on the machine hosting cryosparc_master must exactly match the path on machines hosting cryosparc_worker

• The Anaconda installation directory must be accessible by the CryoSPARC Linux user account with the required permissions for executing the topaz binary

Use a previously-installed Anaconda Python (3.6+), or install new one. Alternatively, Miniforge3 provides the required conda functionality.

Create and activate aconda environment named topaz environment with the following commands:

conda create -n topaz python=3.6
conda activate topaz  # changes to the topaz conda environment

For Topaz 0.2.5, thecondaenvironment should be configured with Python 3.6, regardless of the versions of other Python installations on the system.

Once the conda environment has been created and activated, run the following command to install Topaz:

conda install topaz=0.2.5 mkl=2024.0.0 -c tbepler -c pytorch -c conda-forge

After running the preceding command, with the condaenvironment still active, the command

topaz --version

should print TOPAZ 0.2.5a to the terminal.

Find the Topaz Executable Path

Once Topaz is installed and the conda environment is active in your current shell, enter the following command to determine the full path to the topaz binary:

which topaz

The output should look similar to this:

/home/cryosparcuser/anaconda3/envs/topaz/bin/topaz

In the CryoSPARC interface, specify this as the value for the "Path to Topaz executable" parameter:

If using this path results in errors when running a Topaz job (often with a message such as "topaz did not produce valid output"), try aliasing Topaz with a shell script that also activates the correct conda environment. This process is described in the next section.

(Optional) Create a topaz.sh wrapper script

CryoSPARC has its own conda installation and environment, which may conflict with the Topaz environment. To prevent this, create a shell script that deactivates the CryoSPARC environment and activates the topaz one.

Create a topaz.sh file in a well known location such as the home directory (e.g., ~/topaz.sh). Add the following contents, making the noted substitutions:

#!/usr/bin/env bash
if command -v conda > /dev/null 2>&1; then
    conda deactivate > /dev/null 2>&1 || true  # ignore any errors
    conda deactivate > /dev/null 2>&1 || true  # ignore any errors
fi
unset _CE_CONDA
unset CONDA_DEFAULT_ENV
unset CONDA_EXE
unset CONDA_PREFIX
unset CONDA_PROMPT_MODIFIER
unset CONDA_PYTHON_EXE
unset CONDA_SHLVL
unset PYTHONPATH
unset LD_PRELOAD
unset LD_LIBRARY_PATH

source $HOME/anaconda3/etc/profile.d/conda.sh
conda activate topaz
exec topaz $@

• Substitute $HOME/anaconda3 on line 17 with the Anaconda or Miniforge installation directory.

Make this file executable by the CryoSPARC user from the command line

chmod +x topaz.sh

In the CryoSPARC interface, specify the full path to topaz.sh as the "Path to Topaz executable" parameter:

(Optional) Set Topaz executable path as project-level default parameter

To avoid having to locate and set the Topaz executable path when building every Topaz job, in v4.0.2 onwards you can set a project-level default that will apply to all newly created Topaz jobs.

Navigate to the projects view, select a project and choose the Topaz executable path under the 'Project Level Parameters' module within the sidebar details panel:

You can use the input field or file browser to enter a path to the Topaz executable.

Verify Topaz Installation

Log out of the current command shell and log in again to ensure no conda environment is active. Run the following commands to verify that the Topaz Installation is working correctly. These are adapted from the Topaz Quick start guide. Note the substitutions below.

wget http://bergerlab-downloads.csail.mit.edu/topaz/topaz-tutorial-data.tar.gz
tar -xzvf topaz-tutorial-data.tar.gz
mkdir -p data/EMPIAR-10025/processed
mkdir -p data/EMPIAR-10025/processed/micrographs
/path/to/topaz preprocess -v -s 8 -o data/EMPIAR-10025/processed/micrographs/ data/EMPIAR-10025/rawdata/micrographs/*.mrc
/path/to/topaz convert -s 8 -o data/EMPIAR-10025/processed/particles.txt data/EMPIAR-10025/rawdata/particles.txt

• Substitute /path/to/topaz with the full path to be used for CryoSPARC jobs

• For the preprocess command, specify one of

  • --device n where n is a specific GPU number to run the test on

  • --num-workers n where n is the number of processes to use

Job Types

Job: Topaz Train and Job: Topaz Cross Validation

Job: Topaz Extract

Job: Topaz Denoise

Topaz Deep Picking Tutorials

T20S Proteasome: Topaz Particle Picking Tutorial

T20S Proteasome: Topaz Micrograph Denoising Tutorial