◎aibioatlasExplore · Compare · Discover
Home›Developer Hub & Python SDK
API & Developer Tools

BioAtlas Developer Hub & Python SDK

Integrate the world’s largest index of biological AI foundation models, structural benchmarks, and automated inference endpoints directly into your pipelines with a single line of code.

REST API Reference

⚡ 60-Second Quickstart

pip install bioatlas
import bioatlas

# Initialize client with your API key
client = bioatlas.Client(api_key="YOUR_BIOATLAS_API_KEY")

# 1. Query verified models by modality and benchmark score
top_models = client.models.list(
    modality="protein-ligand",
    min_casp_score=85.0,
    license="open-source"
)
for model in top_models:
    print(f"{model.name} ({model.company}): CASP15={model.specs.casp15}")

# 2. Run cloud inference on AlphaFold 3 / ESM-3 endpoint
result = client.predict(
    model="alphafold-3",
    sequence="EVQLVESGGGLVQPGGSLRLSCAASGFTFSSYAMSWVRQAPGKGLEWVSAISGSGGSTYYADSVKGRFTISRDNSKNTLYLQMNSLRAEDTAVYYCAK...",
    include_pae=True
)

# 3. Export atomic coordinates to PDB file
result.save_pdb("predicted_structure.pdb")
print(f"Mean pLDDT: {result.mean_plddt:.2f}")

📖 Core REST API Endpoints

All requests require an `Authorization: Bearer <API_KEY>` header. Base URL: `https://api.bioatlas.org/v1`.

GET/v1/models

List, filter, and paginate through verified biological AI models with specifications, benchmark rankings, and license information.

QUERY PARAMETERS:
category (string), modality (string), license (open-source | proprietary), limit (int, max 100).
POST/v1/predict

Submit asynchronous inference tasks for macromolecular structure prediction (AlphaFold 3, Chai-1, ESM-3). Returns task ID and webhook notification URL.

REQUEST BODY JSON:
model (string), sequence (string), seed (int, optional), num_recycle (int, default 3).
GET/v1/benchmarks/:id

Retrieve full benchmark leaderboard data for CASP15, CASP16, CAMEO, MoleculeNet, or ProteinGym with per-target breakdown scores.