alphafold — protein structure prediction
DeepMind’s AlphaFold — state-of-the-art protein structure prediction from sequence alone. The image ships AlphaFold + the genetic / template databases it needs at inference time.
Image tag
docker.io/manvarharsh/alphafold:cuda12
What’s in this image
- Base:
nvidia/cuda:12.4.1-cudnn-devel-ubuntu22.04 - Python 3.10 (conda)
- AlphaFold inference pipeline
- HMMER, HHsuite, Kalign for MSA / template search
- JupyterHub with Podstack authenticator
- OpenSSH server
Default ports
| Port | Service |
|---|---|
| 22 | SSH |
| 8000 | JupyterHub |
Use cases
- Predicting 3D structure from a protein sequence
- Multimer predictions
- High-throughput structural biology screens
- Reproducible MSA + structure pipelines
Environment variables
| Variable | Description |
|---|---|
ENABLE_SSH | Enable SSH server |
ENABLE_JUPYTERHUB | Enable JupyterHub on port 8000 |
PODSTACK_API_URL | Backend URL for JupyterHub token validation |
SSH_PUBLIC_KEY | Public key for SSH |
Persistence
Mount at /data. AlphaFold’s reference databases are large (several TB) — point the genetic-database flags at a mounted NFS volume under /data/alphafold-db/.