lucidrains/alphagenome

Implementation of AlphaGenome, Deepmind's updated genomic attention model

55
/ 100
Established

Built on a UNet-Transformer architecture with multi-scale attention mechanisms, it produces hierarchical embeddings at 1bp, 128bp, and pairwise resolution levels for genomic sequences. Supports multi-organism training (human, mouse, etc.), modular prediction heads for chromatin tracks, 3D contacts, and splicing effects, with compatibility for loading official JAX pretrained weights and TFRecord dataset pipelines.

No Package No Dependents
Maintenance 13 / 25
Adoption 9 / 25
Maturity 15 / 25
Community 18 / 25

How are scores calculated?

Stars

97

Forks

17

Language

Jupyter Notebook

License

MIT

Last pushed

Mar 25, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/lucidrains/alphagenome"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.