ipd/proteinmpnn

Pull and run ipd/proteinmpnn using Docker (this will download the full model and run it in your local environment)

Get API KeyCopy Code
$ docker login nvcr.io
Username: $oauthtoken
Password: <PASTE_API_KEY_HERE>

Start NIM

1. Export NGC_CLI_API_KEY variable.

Copy Code
export NGC_CLI_API_KEY=<your personal NGC key>

2. NIM container automatically downloads models. To save time and bandwidth it is recommended to provide local cache directory. This way NIM will be able to reuse already downloaded models. Execute following command to setup cache directory.

Copy Code
export LOCAL_NIM_CACHE=~/.cache/nim
mkdir -p "$LOCAL_NIM_CACHE"
sudo chmod 0777 -R "$LOCAL_NIM_CACHE"

3. Run the NIM container with the following commands.

Copy Code
docker run -it \
    --runtime=nvidia \
    --gpus='"device=0"' \
    -p 8000:8000 \
    -e NGC_CLI_API_KEY \
    -v "$LOCAL_NIM_CACHE":/home/nvs/.cache/nim \
    nvcr.io/nim/ipd/proteinmpnn:1.0.0

This command will start the NIM container and expose port 8000 for the user to interact with the NIM.

4. Open a new terminal, leaving the terminal open with the just launched service. In the new terminal, wait until the health check end point returns {"status":"ready"} before proceeding. This may take a couple of minutes. You can use the following command to query the health check.

Copy Code
curl http://localhost:8000/v1/health/ready

Python client example

1. Save following Python example to a file named nim_client.py.

Copy Code
#!/usr/bin/env python3
import requests
import os
import json
from pathlib import Path

def get_reduced_pdb():
    pdb = Path("1R42.pdb")
    if not pdb.exists():
        pdb.write_text(requests.get(f"https://files.rcsb.org/download/{pdb}").text)
    lines = filter(lambda line: line.startswith("ATOM"), pdb.read_text().split("\n"))
    return "\n".join(list(lines)[:200])

r = requests.post(
    url="http://localhost:8000/biology/ipd/proteinmpnn/predict",
    json={
        "input_pdb": get_reduced_pdb(),
        "ca_only": False,
        "use_soluble_model": False,
        "sampling_temp": [0.1],
    },
)
print(r, "Saving to output.fa:\n", r.text[:200], "...")
Path("output.fa").write_text(json.loads(r.text)["mfasta"])

2. Execute the example.

Copy Code
chmod +x nim_client.py

./nim_client.py

3. The NIM saves results to the output.fa file in Multi-FASTA format. You can quickly view the file using the following command.

Copy Code
cat output.fa

Shell client example

1. Save the following Shell example to a file named nim_client.sh.

Copy Code
#!/usr/bin/env bash
set -e

URL=http://localhost:8000/biology/ipd/proteinmpnn/predict

if [ ! -e 1R42.pdb ]; then curl -O https://files.rcsb.org/download/1R42.pdb; fi

pdb=$(cat 1R42.pdb | grep ^ATOM | head -n 200 | awk '{printf "%s\\n", $0}')

request='{
 "input_pdb": "'"$pdb"'",
 "ca_only": false,
 "use_soluble_model": false,
 "sampling_temp": [0.1]
}'
curl -H 'Content-Type: application/json' \
     -d "$request" "$URL"

2. Execute the example.

Copy Code
chmod +x nim_client.sh

./nim_client.sh

3. The NIM displays the results to the terminal in JSON format. You can configure the NIM to display the results, scores, and probabilities in the Multi-FASTA output format.

For more details on getting started with this NIM, visit the NVIDIA NIM Docs.