RFDiffusion - HackMD

# RFDiffusion https://github.com/RosettaCommons/RFdiffusion.git ## Setup ```bash git clone https://github.com/RosettaCommons/RFdiffusion.git cd RFdiffusion mkdir models && cd models wget http://files.ipd.uw.edu/pub/RFdiffusion/6f5902ac237024bdd0c176cb93063dc4/Base_ckpt.pt wget http://files.ipd.uw.edu/pub/RFdiffusion/e29311f6f1bf1af907f9ef9f44b8328b/Complex_base_ckpt.pt wget http://files.ipd.uw.edu/pub/RFdiffusion/60f09a193fb5e5ccdc4980417708dbab/Complex_Fold_base_ckpt.pt wget http://files.ipd.uw.edu/pub/RFdiffusion/74f51cfb8b440f50d70878e05361d8f0/InpaintSeq_ckpt.pt wget http://files.ipd.uw.edu/pub/RFdiffusion/76d00716416567174cdb7ca96e208296/InpaintSeq_Fold_ckpt.pt wget http://files.ipd.uw.edu/pub/RFdiffusion/5532d2e1f3a4738decd58b19d633b3c3/ActiveSite_ckpt.pt wget http://files.ipd.uw.edu/pub/RFdiffusion/12fc204edeae5b57713c5ad7dcb97d39/Base_epoch8_ckpt.pt #Optional: wget http://files.ipd.uw.edu/pub/RFdiffusion/f572d396fae9206628714fb2ce00f72e/Complex_beta_ckpt.pt # original structure prediction weights wget http://files.ipd.uw.edu/pub/RFdiffusion/1befcb9b28e2f778f53d47f18b7597fa/RF_structure_prediction_weights.pt cd .. mkdir inputs && cd inputs wget https://files.rcsb.org/view/5TPN.pdb # ugly fix for https://github.com/RosettaCommons/RFdiffusion/issues/79 cd .. mkdir shedules ``` ## Building the Docker image and converting to a Singularity container One needs a computer (preferably one running Linux OS) with docker and Singularity/Apptainer installed (one can use gitpod workspace to install Apptainer). Alternatively, one can build the docker container, push to a repository and the pull with singularity on Rackham. 1. Follow the original instructions to build the docker image https://github.com/RosettaCommons/RFdiffusion#docker 2. Convert to Singularity container - on the same computer where docker image was build ```bash sudo singularity build rfdiffusion.sif docker-daemon://rfdiffusion:latest ``` - Via DockerHub (*as example*) ```bash singularity pull docker://xxxx/rfdiffusion:latest ``` ## Run ```bash cd RFdiffusion singularity exec --nv \ --env TF_FORCE_UNIFIED_MEMORY=1,XLA_PYTHON_CLIENT_MEM_FRACTION=4.0,OPENMM_CPU_THREADS=10,HYDRA_FULL_ERROR=1 \ -B $PWD/schedules:/usr/local/lib/python3.9/dist-packages/schedules \ ../rfdiffusion.sif \ python3.9 /app/RFdiffusion/scripts/run_inference.py \ inference.output_prefix=outputs/motifscaffolding \ inference.model_directory_path=$PWD/models \ inference.input_pdb=$PWD/inputs/5TPN.pdb \ inference.num_designs=3 \ 'contigmap.contigs=[10-40/A163-181/10-40]' ``` ``` [2023-11-14 11:36:31,738][__main__][INFO] - Found GPU with device_name Quadro K2200. Will run RFdiffusion on Quadro K2200 Reading models from models [2023-11-14 11:36:31,739][rfdiffusion.inference.model_runners][INFO] - Reading checkpoint from models/Base_ckpt.pt This is inf_conf.ckpt_path models/Base_ckpt.pt Assembling -model, -diffuser and -preprocess configs from checkpoint USING MODEL CONFIG: self._conf[model][n_extra_block] = 4 USING MODEL CONFIG: self._conf[model][n_main_block] = 32 USING MODEL CONFIG: self._conf[model][n_ref_block] = 4 USING MODEL CONFIG: self._conf[model][d_msa] = 256 USING MODEL CONFIG: self._conf[model][d_msa_full] = 64 USING MODEL CONFIG: self._conf[model][d_pair] = 128 USING MODEL CONFIG: self._conf[model][d_templ] = 64 USING MODEL CONFIG: self._conf[model][n_head_msa] = 8 USING MODEL CONFIG: self._conf[model][n_head_pair] = 4 USING MODEL CONFIG: self._conf[model][n_head_templ] = 4 USING MODEL CONFIG: self._conf[model][d_hidden] = 32 USING MODEL CONFIG: self._conf[model][d_hidden_templ] = 32 USING MODEL CONFIG: self._conf[model][p_drop] = 0.15 USING MODEL CONFIG: self._conf[model][SE3_param_full] = {'num_layers': 1, 'num_channels': 32, 'num_degrees': 2, 'n_heads': 4, 'div': 4, 'l0_in_features': 8, 'l0_out_features': 8, 'l1_in_features': 3, 'l1_out_features': 2, 'num_edge_features': 32} USING MODEL CONFIG: self._conf[model][SE3_param_topk] = {'num_layers': 1, 'num_channels': 32, 'num_degrees': 2, 'n_heads': 4, 'div': 4, 'l0_in_features': 64, 'l0_out_features': 64, 'l1_in_features': 3, 'l1_out_features': 2, 'num_edge_features': 64} USING MODEL CONFIG: self._conf[model][freeze_track_motif] = False USING MODEL CONFIG: self._conf[model][use_motif_timestep] = True USING MODEL CONFIG: self._conf[diffuser][T] = 50 USING MODEL CONFIG: self._conf[diffuser][b_0] = 0.01 USING MODEL CONFIG: self._conf[diffuser][b_T] = 0.07 USING MODEL CONFIG: self._conf[diffuser][schedule_type] = linear USING MODEL CONFIG: self._conf[diffuser][so3_type] = igso3 USING MODEL CONFIG: self._conf[diffuser][crd_scale] = 0.25 USING MODEL CONFIG: self._conf[diffuser][so3_schedule_type] = linear USING MODEL CONFIG: self._conf[diffuser][min_b] = 1.5 USING MODEL CONFIG: self._conf[diffuser][max_b] = 2.5 USING MODEL CONFIG: self._conf[diffuser][min_sigma] = 0.02 USING MODEL CONFIG: self._conf[diffuser][max_sigma] = 1.5 USING MODEL CONFIG: self._conf[preprocess][sidechain_input] = False USING MODEL CONFIG: self._conf[preprocess][motif_sidechain_input] = True USING MODEL CONFIG: self._conf[preprocess][d_t1d] = 22 USING MODEL CONFIG: self._conf[preprocess][d_t2d] = 44 USING MODEL CONFIG: self._conf[preprocess][prob_self_cond] = 0.5 USING MODEL CONFIG: self._conf[preprocess][str_self_cond] = True USING MODEL CONFIG: self._conf[preprocess][predict_previous] = False [2023-11-14 11:36:52,768][rfdiffusion.inference.model_runners][INFO] - Loading checkpoint. [2023-11-14 11:36:58,359][rfdiffusion.diffusion][INFO] - Calculating IGSO3. Successful diffuser __init__ [2023-11-14 11:37:05,446][__main__][INFO] - Making design outputs/motifscaffolding_0 [2023-11-14 11:37:05,685][rfdiffusion.inference.model_runners][INFO] - Using contig: ['10-40/A163-181/10-40'] With this beta schedule (linear schedule, beta_0 = 0.04, beta_T = 0.28), alpha_bar_T = 0.00013696048699785024 [2023-11-14 11:37:05,856][rfdiffusion.inference.model_runners][INFO] - Sequence init: ----------------------------------------EVNKIKSALLSTNKAVVSL----------------------- /usr/local/lib/python3.9/dist-packages/e3nn/o3/_spherical_harmonics.py:82: UserWarning: FALLBACK path has been taken inside: compileCudaFusionGroup. This is an indication that codegen Failed for some reason. To debug try disable codegen fallback path via setting the env variable `export PYTORCH_NVFUSER_DISABLE=fallback` To report the issue, try enable logging via setting the envvariable ` export PYTORCH_JIT_LOG_LEVEL=manager.cpp` (Triggered internally at ../torch/csrc/jit/codegen/cuda/manager.cpp:237.) sh = _spherical_harmonics(self._lmax, x[..., 0], x[..., 1], x[..., 2]) [2023-11-14 11:37:28,815][rfdiffusion.inference.utils][INFO] - Sampled motif RMSD: 0.32 [2023-11-14 11:37:28,839][rfdiffusion.inference.model_runners][INFO] - Timestep 50, input to next step: ----------------------------------------EVNKIKSALLSTNKAVVSL----------------------- [2023-11-14 11:37:32,092][rfdiffusion.inference.utils][INFO] - Sampled motif RMSD: 0.30 [2023-11-14 11:37:32,099][rfdiffusion.inference.model_runners][INFO] - Timestep 49, input to next step: ----------------------------------------EVNKIKSALLSTNKAVVSL----------------------- [2023-11-14 11:37:34,088][rfdiffusion.inference.utils][INFO] - Sampled motif RMSD: 0.31 [2023-11-14 11:37:34,095][rfdiffusion.inference.model_runners][INFO] - Timestep 48, input to next step: ----------------------------------------EVNKIKSALLSTNKAVVSL----------------------- [2023-11-14 11:37:36,091][rfdiffusion.inference.utils][INFO] - Sampled motif RMSD: 0.31 [2023-11-14 11:37:36,096][rfdiffusion.inference.model_runners][INFO] - Timestep 47, input to next step: ----------------------------------------EVNKIKSALLSTNKAVVSL----------------------- [2023-11-14 11:37:38,077][rfdiffusion.inference.utils][INFO] - Sampled motif RMSD: 0.30 [2023-11-14 11:37:38,085][rfdiffusion.inference.model_runners][INFO] - Timestep 46, input to next step: ----------------------------------------EVNKIKSALLSTNKAVVSL----------------------- [2023-11-14 11:37:40,025][rfdiffusion.inference.utils][INFO] - Sampled motif RMSD: 0.27 [2023-11-14 11:37:40,031][rfdiffusion.inference.model_runners][INFO] - Timestep 45, input to next step: ----------------------------------------EVNKIKSALLSTNKAVVSL----------------------- [2023-11-14 11:37:41,965][rfdiffusion.inference.utils][INFO] - Sampled motif RMSD: 0.27 [2023-11-14 11:37:41,970][rfdiffusion.inference.model_runners][INFO] - Timestep 44, input to next step: ----------------------------------------EVNKIKSALLSTNKAVVSL----------------------- [2023-11-14 11:37:43,908][rfdiffusion.inference.utils][INFO] - Sampled motif RMSD: 0.26 [2023-11-14 11:37:43,916][rfdiffusion.inference.model_runners][INFO] - Timestep 43, input to next step: ----------------------------------------EVNKIKSALLSTNKAVVSL----------------------- [2023-11-14 11:37:45,841][rfdiffusion.inference.utils][INFO] - Sampled motif RMSD: 0.26 [2023-11-14 11:37:45,846][rfdiffusion.inference.model_runners][INFO] - Timestep 42, input to next step: ----------------------------------------EVNKIKSALLSTNKAVVSL----------------------- [2023-11-14 11:37:47,784][rfdiffusion.inference.utils][INFO] - Sampled motif RMSD: 0.25 [2023-11-14 11:37:47,793][rfdiffusion.inference.model_runners][INFO] - Timestep 41, input to next step: ----------------------------------------EVNKIKSALLSTNKAVVSL----------------------- ... [2023-11-14 11:41:05,372][rfdiffusion.inference.utils][INFO] - Sampled motif RMSD: 0.28 [2023-11-14 11:41:05,376][rfdiffusion.inference.model_runners][INFO] - Timestep 3, input to next step: -------------EVNKIKSALLSTNKAVVSL-------------------- [2023-11-14 11:41:06,335][rfdiffusion.inference.utils][INFO] - Sampled motif RMSD: 0.28 [2023-11-14 11:41:06,340][rfdiffusion.inference.model_runners][INFO] - Timestep 2, input to next step: -------------EVNKIKSALLSTNKAVVSL-------------------- [2023-11-14 11:41:08,232][__main__][INFO] - Finished design in 0.83 minutes ``` ## Contacts: - [Pavlin Mitev](https://katalog.uu.se/profile/?id=N3-1425) - [UPPMAX](https://www.uppmax.uu.se/) - [AE@UPPMAX - related documentation](/8sqXISVRRquPDSw9o1DizQ) ![](https://live.webb.uu.se/digitalAssets/207/c_207717-l_3-k_bg-city.png) ###### tags: `UPPMAX`

Read more

Installing Python modules on [Bianca] - UPPMAX

R.ROSSETTA - Singularity container

conda and mamba on Rackham@UPPMAX, Dardel@PDC

pb-16S-nf on Bianca