Quick reference

• The official ProteinMPNN docker image.

• Maintained by: openEuler CloudNative SIG.

• Where to get help: openEuler CloudNative SIG, openEuler.

ProteinMPNN | openEuler

ProteinMPNN is a deep learning-based protein sequence design method. It generates protein sequences that fold into given backbone structures with high accuracy.

Learn more on ProteinMPNN GitHub.

Supported tags and respective Dockerfile links

The tag of each proteinmpnn docker image is consist of the version of proteinmpnn and the version of basic image. The details are as follows

Tag	Currently	Architectures
1.0.1-oe2403sp3	ProteinMPNN 1.0.1 on openEuler 24.03-LTS-SP3	amd64, arm64

Usage

In this usage, users can select the corresponding {Tag} and container startup options based on their requirements.

• Pull the openeuler/proteinmpnn image from docker

  docker pull openeuler/proteinmpnn:{Tag}
  

• Start a ProteinMPNN instance

  docker run -it --rm openeuler/proteinmpnn:{Tag} bash
  

• Run a simple example

  # Inside the container
  cd /opt/ProteinMPNN
  
  # Step 1: Parse PDB files to jsonl format
  python3 helper_scripts/parse_multiple_chains.py \
  	--input_path inputs/PDB_monomers/pdbs/ \
  	--output_path parsed_pdbs.jsonl
  
  # Step 2: Run protein sequence design
  python3 protein_mpnn_run.py \
  	--jsonl_path parsed_pdbs.jsonl \
  	--out_folder outputs/ \
  	--num_seq_per_target 2 \
  	--sampling_temp "0.1"
  

• Design with custom PDB files

  # Mount local directory and run
  docker run -it --rm -v /path/to/pdbs:/pdbs openeuler/proteinmpnn:{Tag} bash
  
  # Inside the container, parse and design
  cd /opt/ProteinMPNN
  python3 helper_scripts/parse_multiple_chains.py \
  	--input_path /pdbs/ \
  	--output_path /pdbs/parsed_pdbs.jsonl
  
  python3 protein_mpnn_run.py \
  	--jsonl_path /pdbs/parsed_pdbs.jsonl \
  	--out_folder /pdbs/outputs/ \
  	--num_seq_per_target 10
  

Question and answering

If you have any questions or want to use some special features, please submit an issue or a pull request on openeuler-docker-images.