![[PukiWiki]](image/picc.png "[PukiWiki]")
本家様 https://biopipelines.readthedocs.io/en/latest/
github https://github.com/locbp-uzh/biopipelines
git clone https://github.com/yyuu/pyenv.git /apps/pyenv
export PYENV_ROOT=/apps/pyenv
export PATH=$PYENV_ROOT/bin:$PATH
pyenv install miniforge3-26.3.2-1
source /apps/pyenv/versions/miniforge3-26.3.2-1/etc/profile.d/conda.sh
conda update conda
(既に環境があるなら)
source /apps/pyenv/versions/miniforge3-26.3.2-1/etc/profile.d/conda.shroot@ubuntu24:~# grep VERSION /etc/os-release
VERSION_ID="24.04"
VERSION="24.04.4 LTS (Noble Numbat)"
VERSION_CODENAME=noble
root@ubuntu24:~#
root@ubuntu24:~# cat /proc/driver/nvidia/version
NVRM version: NVIDIA UNIX Open Kernel Module for x86_64 595.71.05 Release Build (dvs-builder@U22-I3-G08-03-1) Fri Apr 24 06:42:30 UTC 2026
GCC version: gcc version 13.3.0 (Ubuntu 13.3.0-6ubuntu2~24.04.1)
root@ubuntu24:~# nvidia-smi -L
GPU 0: NVIDIA RTX PRO 2000 Blackwell (UUID: GPU-40660e37-0d35-d4f4-294a-eee3fe83049e)
root@ubuntu24:~#
root@ubuntu24:~# ls /usr/local/cuda-12.9/
bin extras gds-12.9 lib64 nsight-compute-2025.2.1 nsight-systems-2025.1.3 nvvm src version.json
compute-sanitizer gds include libnvvp nsightee_plugins nvml share targets
root@ubuntu24:~#root@ubuntu24:~# cd /apps/
root@ubuntu24:/apps# git clone https://github.com/locbp-uzh/biopipelines.git
root@ubuntu24:/apps# cd biopipelines
root@ubuntu24:/apps/biopipelines# ls
biopipelines config.local.yaml environments ligands my_pipelines pipe_scripts renderers run tables
config.cluster.yaml CONTRIBUTING.md example_pipelines llm outputs pyproject.toml requirements-docs.txt sequences tests
config.colab.yaml docs LICENSE mkdocs.yml pdbs README.md resubmit submit versions
root@ubuntu24:/apps/biopipelines# ls environments/
admet_ai.pip.txt dssp.yaml frame2seq.yaml placer.yaml ProteinEnv.colab.yaml
admet_ai.yaml dynamicbind.colab.yaml gems.colab.yaml plip.colab.yaml rbs_designer.cluster.yaml
af2bind.yaml dynamicbind.pip.1.txt gems.yaml plip.pip.1.txt rbs_designer.colab.yaml
Aggrescan3D.colab.yaml dynamicbind.pip.2.txt gnina.yaml plip.pip.colab.1.txt reduce.colab.yaml
Aggrescan3D.yaml dynamicbind.pip.colab.1.txt ligandmpnn_env.pip.cluster.txt plip.yaml reduce.yaml
apbs.colab.yaml dynamicbind.pip.colab.2.txt ligandmpnn_env.pip.colab.txt plm_sol.pip.txt rfdiffusion_allatom.pip.txt
apbs.yaml dynamicbind_relax.colab.yaml ligandmpnn_env.yaml plm_sol.yaml rtmscore.colab.yaml
bioemu.yaml dynamicbind_relax.yaml MutationEnv.yaml pocketgen.colab.yaml rtmscore.yaml
biopipelines.yaml dynamicbind.yaml neuralplexer.colab.yaml pocketgen.pip.1.txt SE3nv.cluster.yaml
Boltz2Env.pip.txt esmfold.cluster.yaml neuralplexer.pip.1.txt pocketgen.pip.2.txt SE3nv.colab.yaml
Boltz2Env.yaml esmfold.colab.yaml neuralplexer.pip.2.txt pocketgen.pip.colab.1.txt SE3nv.pip.cluster.txt
boltzgen.pip.txt esmfold.pip.colab.1.txt neuralplexer.pip.3.txt pocketgen.pip.colab.2.txt SE3nv.pip.colab.txt
boltzgen.yaml esmfold.pip.colab.2.txt neuralplexer.pip.colab.1.txt pocketgen.yaml thermompnn.colab.yaml
CABSflex.colab.yaml foundry.pip.txt neuralplexer.pip.colab.2.txt posebusters.pip.txt thermompnn.pip.colab.txt
CABSflex.yaml foundry.yaml neuralplexer.pip.colab.3.txt posebusters.yaml thermompnn.yaml
diffdock.colab.yaml fpocket.colab.yaml neuralplexer.yaml prodigy.colab.yaml vespag.colab.yaml
diffdock.pip.colab.txt fpocket.yaml openmm.colab.yaml prodigy.yaml vespag.pip.colab.txt
diffdock.pip.txt frame2seq.colab.yaml openmm.yaml prolif.colab.yaml vespag.pip.txt
diffdock.yaml frame2seq.pip.colab.txt p2rank.colab.yaml prolif.yaml vespag.yaml
dssp.colab.yaml frame2seq.pip.txt p2rank.yaml ProteinEnv.cluster.yaml xtb.yaml
root@ubuntu24:/apps/biopipelines#
root@ubuntu24:/apps/biopipelines# conda env create -f environments/biopipelines.yaml
root@ubuntu24:/apps/biopipelines# conda activate biopipelines
(biopipelines) root@ubuntu24:/apps/biopipelines#
(biopipelines) root@ubuntu24:/apps/biopipelines# conda list
:
ipython 9.14.0 pyh53cf698_0 conda-forge
:
numpy 2.4.6 py312h33ff503_0 conda-forge
openbabel 3.1.1 py312hbfe4552_9 conda-forge
:
python 3.12.13 hd63d673_0_cpython conda-forge
:
rdkit 2025.09.5 py312h3ecb6ed_0 conda-forge
:
(biopipelines) root@ubuntu24:/apps/biopipelines#
(biopipelines) root@ubuntu24:/apps/biopipelines# conda search "pytorch=2.11=*cuda129*" -c pytorch
Loading channels: done
# Name Version Build Channel
pytorch 2.11.0 cuda129_generic_py310_h331eed0_200 conda-forge
pytorch 2.11.0 cuda129_generic_py311_h49a0b6d_200 conda-forge
pytorch 2.11.0 cuda129_generic_py312_ha5c3ddd_200 conda-forge
pytorch 2.11.0 cuda129_generic_py313_h7c95e44_200 conda-forge
pytorch 2.11.0 cuda129_generic_py314_h117e3a7_200 conda-forge
pytorch 2.11.0 cuda129_mkl_py310_h518f2e7_300 conda-forge
pytorch 2.11.0 cuda129_mkl_py311_h6b7d41d_300 conda-forge
pytorch 2.11.0 cuda129_mkl_py312_hb9da02c_300 conda-forge
pytorch 2.11.0 cuda129_mkl_py313_hd2d7542_300 conda-forge
pytorch 2.11.0 cuda129_mkl_py314_hfa65069_300 conda-forge
(biopipelines) root@ubuntu24:/apps/biopipelines#
(biopipelines) root@ubuntu24:/apps/biopipelines# conda install pytorch=2.11.0=cuda129_generic_py312_ha5c3ddd_200 -c conda-forge
(biopipelines) root@ubuntu24:/apps/biopipelines# python -c "import torch; print(torch.cuda.is_available()); print(torch.cuda.get_device_name(0))"
True
NVIDIA RTX PRO 2000 Blackwell
(biopipelines) root@ubuntu24:/apps/biopipelines#(biopipelines) root@ubuntu24:/apps/biopipelines# conda install pytest
(biopipelines) root@ubuntu24:/apps/biopipelines#
(biopipelines) root@ubuntu24:/apps/biopipelines# pytest tests/
================================================================== test session starts ===================================================================
platform linux -- Python 3.12.13, pytest-9.0.3, pluggy-1.6.0
rootdir: /apps/biopipelines
configfile: pyproject.toml
collected 650 items / 146 deselected / 504 selected
tests/test_bayesian_adjuster.py .... [ 0%]
tests/test_cabsflex_colab.py ...... [ 1%]
tests/test_combinatorics.py ........................ [ 6%]
tests/test_consensus.py .... [ 7%]
tests/test_datastream.py ................................... [ 14%]
tests/test_folders.py .. [ 14%]
tests/test_id_map_utils.py ................................. [ 21%]
tests/test_id_patterns.py ......................................................... [ 32%]
tests/test_internal_and_folder.py .............. [ 35%]
tests/test_ligand.py .............................. [ 41%]
tests/test_load_multiple_folder.py ... [ 42%]
tests/test_missing_propagation.py .......................... [ 47%]
tests/test_mock.py ................................. [ 53%]
tests/test_openbabel.py ..... [ 54%]
tests/test_panda.py .................... [ 58%]
tests/test_parallel.py ....... [ 60%]
tests/test_pdb.py ................................. [ 66%]
tests/test_pipeline_generation.py ..... [ 67%]
tests/test_pocket_validation_wiring.py ....... [ 69%]
tests/test_pool.py .............. [ 71%]
tests/test_provenance.py ................ [ 75%]
tests/test_remap.py .......... [ 76%]
tests/test_resolve_selection_in_sequence.py ..... [ 77%]
tests/test_sequence.py .............................. [ 83%]
tests/test_shell_safety.py ............................................ [ 92%]
tests/test_stream_slicers.py ................. [ 96%]
tests/test_table.py .................... [100%]
===================================================== 504 passed, 146 deselected in 63.13s (0:01:03) =====================================================
(biopipelines) root@ubuntu24:/apps/biopipelines#使いたいアプリがあるなら
conda env update -n biopipelines -f environments/dssp.yamlとかで作った biopipelines 仮想実行環境に組み込むことはできるけど、environments/ にあるものは互いに調整してなくて、組み合わせによっては依存環境で実行できないこともあるそうな....意味ねぇー
実際の計算というかジョブの流し方は
pipeline:
- module: dssp
input_pdb: ./pdbs/1abc.pdb
output_dir: ./outputs/dssp
- module: plip
input_pdb: ./pdbs/1abc.pdb
output_dir: ./outputs/plipとか作って ./run に渡す
jupyterlab とか使った方がいいかも。でも複数のアプリをまとめたconda環境を作る必要がある。機能、アプリ毎にconda環境を作ってもjupyterlab は１つの環境から別の環境に行く際に変数とか消えるので..
