Genomics

SpliceDeep

SpliceDeep is a SYNTHETIC/DEMO genomics model for splice-site & aberrant splicing prediction. Outputs are deterministic demo values — NOT for clinical use. Trained (synthetically) on de-identified sequence data.

Sequence Genoma Institute License: MIT v4.5.9 RTG 3.9 DL 30,350
SYNTHETIC DEMOAll outputs are deterministic demo data — not for clinical use, not real medical inference.

Metrics

MetricValueSplit
PR-AUC0.887Synthetic hold-out
Calibration (ECE)0.042Synthetic
InputsOutputs
Genomic sequence windowSplice delta score

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-026-splicedeep/predict",
    headers={"Authorization": "Bearer <API_KEY>"},
    json={"input": "<genomic sequence window>"})
print(r.json())  # -> synthetic demo prediction

See the API reference · JSON: /api/models/bf-026-splicedeep

Version history

VersionStatusNotes
v4.5.0ArchivedSynthetic release note 0
v4.5.1ArchivedSynthetic release note 1
v4.5.2ArchivedSynthetic release note 2
v4.5.3ArchivedSynthetic release note 3
v4.5.4ArchivedSynthetic release note 4
v4.5.5ArchivedSynthetic release note 5
v4.5.6ArchivedSynthetic release note 6
v4.5.7ArchivedSynthetic release note 7
v4.5.8ArchivedSynthetic release note 8
v4.5.9CurrentSynthetic release note 9

Community contributions: 22 synthetic contributors have submitted evaluations and fine-tunes.