Model details
View repositoryRFdiffusion is a diffusion-based deep learning model that generates novel protein structures and complexes, progressively refining structures to create diverse, high-quality 3D protein models. It takes structural constraints (e.g., partial protein structures in PDB format) as input and outputs generated 3D protein structures, supporting tasks like protein scaffolding and binder design tailored to specific target molecules.
Call BioNeMo RFdiffusion on Baseten using the NIM protein design route.
Input
1import os
2import requests
3
4MODEL_ID = os.environ["BASETEN_MODEL_ID"]
5BASETEN_API_KEY = os.environ["BASETEN_API_KEY"]
6BASETEN_URL = (
7 f"https://model-{MODEL_ID}.api.baseten.co/environments/production/sync"
8 "/biology/ipd/rfdiffusion/generate"
9)
10
11pdb_text = requests.get("https://files.rcsb.org/download/1R42.pdb", timeout=30).text
12atom_lines = [line for line in pdb_text.splitlines() if line.startswith("ATOM")][:400]
13input_pdb = "\n".join(atom_lines)
14
15payload = {
16 "input_pdb": input_pdb,
17 "contigs": "A20-60/0 50-100",
18 "hotspot_res": ["A50", "A51", "A52", "A53", "A54"],
19 "diffusion_steps": 15,
20}
21
22response = requests.post(
23 BASETEN_URL,
24 headers={
25 "Authorization": f"Bearer {BASETEN_API_KEY}",
26 "Content-Type": "application/json",
27 },
28 json=payload,
29 timeout=300,
30)
31response.raise_for_status()
32result = response.json()
33print(result)
34JSON output
1{
2 "output_pdb": "<generated protein backbone PDB>",
3 "elapsed_ms": 123
4}
