AlphaLink2

PaperCodeNeed >20GB

Protein complex structure prediction from crosslinking mass-spectrometry restraints (XL-MS)

Readiness

Readiness: Need >20GB
Smoke: passed
Artifacts: passed
Production: deferred_unifold_alphafold_databases_gpu_stack
Storage: >20GB
Strategy: shared python312_slim.sif + tool ext3 overlay
Next: Defer to large database phase: build shared Uni-Fold/AlphaFold GPU container, stage model parameters and reduced/full databases, run example_data/H1142 through Slurm, then add runner adapter/artifacts.

SOP main-thread triage staged upstream AlphaLink2 revision 3bc0cd2965aea1c458f0ef2629c8af788b66fb7b at /media/nik/seagate_nik/bio_server/tools/alphalink2/source; source is about 73M and includes example_data. README requires Uni-Core/Uni-Fold/AlphaFold runtime, OpenMM/pdbfixer, HMMER/HH-suite/Kalign, AlphaLink2 model weights from Zenodo, and MSA/template databases; full database script can require up to 3TB. No Slurm benchmark run because parameters/databases/runtime are not staged.