Drug Discovery

DockFormer-Bind

DockFormer-Bind is a SYNTHETIC/DEMO drug discovery model for protein-ligand binding affinity prediction. Outputs are deterministic demo values — NOT for clinical use. Trained (synthetically) on de-identified sequence data.

Sequence PathAI Community License: CC-BY-4.0 v4.4.3 RTG 4.5 DL 399,277
SYNTHETIC DEMOAll outputs are deterministic demo data — not for clinical use, not real medical inference.

Metrics

MetricValueSplit
Pearson r0.948Synthetic hold-out
Calibration (ECE)0.031Synthetic
InputsOutputs
SMILES + protein sequencepKd + pose confidence

Run a synthetic prediction

Returns a deterministic synthetic result — clearly labeled, not for clinical use.

API integration

Call this model from your pipeline (demo — no real key required):

import requests
r = requests.post("https://api.biofoundry.ai/v1/models/bf-013-dockformer-bind/predict",
    headers={"Authorization": "Bearer <API_KEY>"},
    json={"input": "<smiles + protein sequence>"})
print(r.json())  # -> synthetic demo prediction

See the API reference · JSON: /api/models/bf-013-dockformer-bind

Version history

VersionStatusNotes
v4.4.0ArchivedSynthetic release note 0
v4.4.1ArchivedSynthetic release note 1
v4.4.2ArchivedSynthetic release note 2
v4.4.3CurrentSynthetic release note 3

Community contributions: 82 synthetic contributors have submitted evaluations and fine-tunes.