Protein Folding

MembraneFold-TM

MembraneFold-TM is a SYNTHETIC/DEMO protein folding model for transmembrane topology prediction. Outputs are deterministic demo values — NOT for clinical use. Trained (synthetically) on de-identified sequence data.

Sequence DeepCell AI License: CC-BY-4.0 v3.4.6 RTG 3.9 DL 418,401
SYNTHETIC DEMOAll outputs are deterministic demo data — not for clinical use, not real medical inference.

Metrics

MetricValueSplit
Q30.825Synthetic hold-out
Calibration (ECE)0.037Synthetic
InputsOutputs
SequenceTopology + helices

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-024-membranefold-tm/predict",
    headers={"Authorization": "Bearer <API_KEY>"},
    json={"input": "<sequence>"})
print(r.json())  # -> synthetic demo prediction

See the API reference · JSON: /api/models/bf-024-membranefold-tm

Version history

VersionStatusNotes
v3.4.0ArchivedSynthetic release note 0
v3.4.1ArchivedSynthetic release note 1
v3.4.2ArchivedSynthetic release note 2
v3.4.3ArchivedSynthetic release note 3
v3.4.4ArchivedSynthetic release note 4
v3.4.5ArchivedSynthetic release note 5
v3.4.6CurrentSynthetic release note 6

Community contributions: 64 synthetic contributors have submitted evaluations and fine-tunes.