PLACER

5-30 minutesPaperCodeNeed >5GB

Protein-ligand docking

Readiness

Readiness: Need >5GB
Smoke: failed
Artifacts: failed
Production: need_gt5gb_placer_cuda121_torch23_dgl24_openbabel_sif_pending
Storage: 5-20GB
Strategy: Dedicated/shared CUDA12.1 PyTorch2.3 + DGL2.4 + e3nn + OpenBabel SIF; repo weights can be reused from source checkout.
Next: Build a dedicated PLACER CUDA 12.1 / PyTorch 2.3 / DGL 2.4 / OpenBabel / e3nn runtime SIF from envs/placer_env_lite.yml, run bundled examples/inputs plus ligand example through Slurm, verify ranked PDB/CSV outputs, then wire adapter.

No obvious large model/database dependency identified; validate upstream docs before install. Worker 4 evidence 2026-07-01: source https://github.com/baker-laboratory/PLACER revision 7dc5563 staged at /media/nik/seagate_nik/bio_server/tools/placer/source, storage 230MB with bundled ~61MB weight. Slurm probe job 263 on GPU partition ran dependency/help probe for 3s but blocked before benchmark: current shared env has torch 2.12.1+cpu/openbabel/numpy and lacks pandas, dgl, e3nn; upstream requires CUDA PyTorch 2.3 plus DGL 2.4 CUDA 12.1. 2026-07-07 16:42 +0800: SOP Production Blocked audit: moved to Need >5GB because PLACER needs a dedicated CUDA PyTorch/DGL/e3nn runtime; the prior GPU probe only reached dependency/help checks and did not run a scientific benchmark. 2026-07-17T01:18:00+08:00 SOP Need >5GB next-10 audit: remains Need >5GB. Runner audit is not_configured; current shared assets do not contain PLACER source/container. Existing notes require CUDA PyTorch 2.3, DGL 2.4 CUDA 12.1, e3nn, and a real benchmark before adapter wiring. 2026-07-18T21:37:01+08:00 SOP Need >5GB batch source-staging: Official repo cloned to backend /media/nik/seagate_nik/bio_server/repos/placer at 7dc5563 (204MB). README confirms weights are included in repo, but runtime needs CUDA PyTorch 2.3, DGL 2.4 CUDA 12.1, OpenBabel, pandas, and e3nn. No runtime/SIF/benchmark yet. 2026-07-18T22:34:12+08:00 SOP Need >5GB continuation batch: No new runtime build completed in this continuation. Source and bundled weights remain staged; next production step remains CUDA12.1 PyTorch2.3 + DGL2.4 + e3nn/OpenBabel SIF and PLACER example benchmark. 2026-07-19T22:38:00+08:00 SOP Need >5GB active install batch: Backend source and bundled weights verified at /media/nik/seagate_nik/bio_server/repos/placer commit 7dc5563 (204MB, weights/PLACER_model_1.pt present). Shared SIF/site and existing Python3.10 mamba env were probed previously: torch/pandas/networkx/numpy are present in some layers, but dgl, e3nn, and openbabel are missing, and PLACER import fails on openbabel. This is an exact runtime/SIF blocker, not a code-link or token blocker.