dsimb/sword2 repository overview

conda env create -f environment.yml
or
mamba env create -f environment.yml

bash install.sh

docker pull dsimb/sword2

# This will mount the local `results` directory inside the Docker image and run the program inside the image as you (local user)
# so that the results generated have good rights.
docker run -it -e LOCAL_UID=$(id -u $USER) -e LOCAL_GID=$(id -g $USER) -v $(pwd)/results:/app/results dsimb/sword2 -p 1jx4 -o results

conda activate sword2

./SWORD2.py -p 1jx4 -o results

./SWORD2.py -u Q76EI6 -o results

./SWORD2.py -m MGYP000936678158 -o results

./SWORD2.py -i ./structure.pdb -o results

./SWORD2.py -i ./structure.pdb -d 2 -o results

./SWORD2.py -p 1jx4 -o results --disable-energy --disable-plots

$ ./SWORD2.py --help
usage: SWORD2.py [-h] [--version] (-u UNIPROT_ID | -m MGNIFY_ID | -p PDB_ID | -i INPUT_FILE) [-c PDB_CHAIN] [-d MODEL] [-x CPU] [-e] -o OUTPUT

SWORD2: SWift and Optimized Recognition of protein Domains.
The SWORD2 partitioning algorithm produces multiple alternative
domain assignments for a given protein structure.
This unique approach handles ambiguous protein structure partitioning,
admitting several solutions. The decomposition of the protein structure
into domains is achieved through the hierarchical clustering of Protein Units,
evolutionarily preserved structural descriptors at the interface between
secondary structures and domains.

options:
  -h, --help            show this help message and exit
  --version             show program's version number and exit
  -u UNIPROT_ID, --uniprot-id UNIPROT_ID
                        AlphaFold Uniprot Accession Id.
  -m MGNIFY_ID, --mgnify-id MGNIFY_ID
                        MGnify Id for the ESM Metagenomic Atlas.
  -p PDB_ID, --pdb-id PDB_ID
                        PDB id to download from the PDB database.
  -i INPUT_FILE, --input-file INPUT_FILE
                        Path to an input PDB or mmCIF file.

optional arguments:
  -c PDB_CHAIN, --pdb-chain PDB_CHAIN
                        PDB chain. Default is A.
  -d MODEL, --model MODEL
                        Model to parse. Especially usefull for NMR files which contain several models. Default is 1.
  -x CPU, --cpu CPU     Number of CPUs to use. Default all (0). Max on this computer is: 40
  -e, --disable-energies
                        Disable the calculation of pseudo-energy of domains and PUs.

required arguments:
  -o OUTPUT, --output OUTPUT
                        Output directory. Results will be generated inside in a dedicated directory named after OUTPUT/PDBID_CHAIN/

$ conda activate sword2
(sword2) $ ./SWORD2.py -p 1jx4 -o results
2024/09/17 16:23:34 INFO     Fetch PDB ID: 1jx4
2024/09/17 16:23:34 DEBUG    Connecting wwPDB FTP server RCSB PDB (USA).
2024/09/17 16:23:36 DEBUG    1jx4 downloaded (results/1jx4_A/1jx4.pdb)
2024/09/17 16:23:36 DEBUG    PDB download via FTP completed (1 downloaded, 0 failed).
2024/09/17 16:23:36 DEBUG    3162 atoms and 1 coordinate set(s) were parsed in 0.05s.
2024/09/17 16:23:36 INFO
2024/09/17 16:23:36 INFO     >>>   Estimated runtime: 50 seconds
2024/09/17 16:23:36 INFO     >>>   Using 40 cpus
2024/09/17 16:23:36 INFO
2024/09/17 16:23:36 INFO     Write a clean version of the PDB: remove non standard residues
2024/09/17 16:23:36 INFO     Launch SWORD
2024/09/17 16:23:49 INFO     Parse SWORD output
2024/09/17 16:23:49 INFO     Calculate pseudo-energies of Domains
2024/09/17 16:23:57 INFO     Write the SWORD results
2024/09/17 16:23:57 INFO     Writing Peeling results
2024/09/17 16:24:01 INFO     Finished writing Peeling results
2024/09/17 16:24:01 INFO     Generate histogram of SWORD2 domains consistency
2024/09/17 16:24:22 INFO     Calculate junctions consistencies
2024/09/17 16:24:22 INFO     Clean and prepare results
2024/09/17 16:24:22 INFO     Results can be found here: results/1jx4_A

import subprocess
from concurrent.futures import ThreadPoolExecutor

# List of 5 PDB codes
pdb_list = ['1jx4', '2c78', '1f5n', '1a8y', '1b89']

# Function to run a SWORD2 job
def run_sword2(pdb_code):
    command = f'./SWORD2.py -p {pdb_code} -o results'
    try:
        subprocess.run(command, shell=True, check=True)
        print(f"Job for {pdb_code} completed successfully.")
    except subprocess.CalledProcessError as e:
        print(f"Error running job for {pdb_code}: {e}")

# Run 5 jobs using ThreadPoolExecutor
def run_sword2_jobs():
    with ThreadPoolExecutor(max_workers=5) as executor:  # Adjust max_workers based on your system's capabilities
        executor.map(run_sword2, pdb_list)

if __name__ == "__main__":
    run_sword2_jobs()