NA-MPNN

1-5 minutesPaperCodeNeed >5GB

Sequence design for protein, RNA, DNA, and mixed-polymer structures; protein-DNA binding specificity prediction

Readiness

Readiness: Need >5GB
Smoke: passed
Artifacts: passed
Production: deferred_prody_openbabel_runtime
Storage: 5-20GB
Strategy: shared PyTorch SIF + tool ext3 overlay; avoid per-tool torch duplication
Next: Build reusable ProDy/OpenBabel-capable PyTorch runtime, rerun 4oqu design example through Slurm, then add runner adapter/artifacts.

SOP agent staged upstream baker-laboratory/NA-MPNN revision aafa32c87b65046a02db152f6e6235a7078e5fef at /media/nik/seagate_nik/bio_server/tools/na-mpnn/source. Source+bundled models about 248M, but shared PyTorch runtime is 6.1G; Slurm job 379 attempted bundled example 4oqu design via shared SIF and failed immediately with ModuleNotFoundError: prody. Evidence log: /media/nik/seagate_nik/bio_server/reports/na-mpnn_sop_evidence_20260701.json.