New docking programs

[DrrDom]

Here everybody may suggest which docking programs would be reasonable to integrate.

I consider integration of rDockand SurfDock (https://github.com/CAODH/SurfDock), but the latter was quite complex and I could not setup it to run.

SurfDock
(implemented)

Another suggestion is Uni-Dock and Uni-Dock2
https://github.com/dptech-corp/Uni-Dock
https://github.com/dptech-corp/Uni-Dock2

PandaDock
https://github.com/pritampanda15/PandaDock

CarsiDock
https://github.com/carbonsilicon-ai/CarsiDock
(implemented)

FIGRDock
https://github.com/fengshikun/FIGRDock

MetalDock
https://github.com/MatthijsHak/MetalDock

[Feriolet]

I could not install SurfDock as well since some of the libraries do not support M1 Mac. Have you tried setting up SurfDock using the script provided here? Intelligent-Drug-Discovery-Lab/SurfDock#19

[DrrDom]

Thank you, I checked the script. It seems that it would be easier to create a container for that program, because some system libraries may be outdated for some users. For example, on our cluster we have a pretty old libc version. The SurfDock developers for some reasons did not make a dockerfile, but there is a contribution from a user Intelligent-Drug-Discovery-Lab/SurfDock#5 (comment). However, it does not use environment.yaml file supplied in the repo, but some custom installation routine. I think a proper dockerfile would be a good contribution to SurfDock and we can adopt it to our needs.

[DrrDom]

I'm also thinking to integrate Boltz2 (https://github.com/jwohlwend/boltz). It has a robust pipeline and interface and I could install it. However, it would require some efforts to adapt it to easydock interface. Boltz2 is not very fast and would be applicable for small datasets.

[Feriolet]

Thank you, I checked the script. It seems that it would be easier to create a container for that program, because some system libraries may be outdated for some users. For example, on our cluster we have a pretty old libc version. The SurfDock developers for some reasons did not make a dockerfile, but there is a contribution from a user CAODH/SurfDock#5 (comment). However, it does not use environment.yaml file supplied in the repo, but some custom installation routine. I think a proper dockerfile would be a good contribution to SurfDock and we can adopt it to our needs.

I saw the environment.yaml and they have a lot of packages and dependencies. How should we install it without breaking incompatibility? I tried the following Dockerfile, but as expected was not able to install the packages properly.

FROM rapidsai/miniforge-cuda:cuda12.1.1-base-ubuntu22.04-py3.10

WORKDIR /app

# Create the environment:
COPY environment.yaml .
RUN conda env create -f environment.yaml

# Make RUN commands use the new environment:
SHELL ["conda", "run", "-n", "SurfDock", "/bin/bash", "-c"]


ENTRYPOINT ["conda", "run", "--no-capture-output", "-n", "SurfDock"]

[Feriolet]

Nevermind I was able to install the dependencies in conda using the same base image, but different architecture. The base image should be based on x86 instead of aarch. However, it still did not work as it said that dimorphite-dl==1.3.2 does not exist

FROM rapidsai/miniforge-cuda:cuda12.1.1-base-ubuntu22.04-py3.10@sha256:70e41823ceb1f0d9c4d8d54bdce945ce643e7b8df19f1c5e6d5bbc96b987b4ec

WORKDIR /app

# Create the environment:
COPY environment.yaml .
RUN conda env create -f environment.yaml

# Make RUN commands use the new environment:
SHELL ["conda", "run", "-n", "SurfDock", "/bin/bash", "-c"]


ENTRYPOINT ["conda", "run", "--no-capture-output", "-n", "SurfDock"]

[Feriolet]

The environment was built successful with the following environment.yaml

name: SurfDock
channels:
  - mx
  - pyg
  - pytorch
  - nvidia
  - defaults
  - conda-forge
dependencies:
  - _libgcc_mutex=0.1=conda_forge
  - _openmp_mutex=4.5=2_kmp_llvm
  - aiohttp=3.9.3=py310h2372a71_1
  - aiosignal=1.3.1=pyhd8ed1ab_0
  - ambertools=22.5=py310hd182041_0
  - annotated-types=0.6.0=pyhd8ed1ab_0
  - anyio=4.3.0=pyhd8ed1ab_0
  - archspec=0.2.3=pyhd8ed1ab_0
  - argon2-cffi=23.1.0=pyhd8ed1ab_0
  - argon2-cffi-bindings=21.2.0=py310h2372a71_4
  - arpack=3.7.0=hdefa2d7_2
  - arrow=1.3.0=pyhd8ed1ab_0
  - asttokens=2.4.1=pyhd8ed1ab_0
  - astunparse=1.6.3=pyhd8ed1ab_0
  - async-lru=2.0.4=pyhd8ed1ab_0
  - async-timeout=4.0.3=pyhd8ed1ab_0
  - attrs=23.2.0=pyh71513ae_0
  - babel=2.13.1=pyhd8ed1ab_0
  - beautifulsoup4=4.12.3=pyha770c72_0
  - biopandas=0.4.1=pyhd8ed1ab_1
  - blas=2.116=mkl
  - blas-devel=3.9.0=16_linux64_mkl
  - bleach=6.1.0=pyhd8ed1ab_0
  - blosc=1.21.5=hc2324a3_1
  - boltons=24.0.0=pyhd8ed1ab_0
  - boost=1.78.0=py310hcb52e73_5
  - boost-cpp=1.78.0=h2c5509c_4
  - brotli=1.1.0=hd590300_1
  - brotli-bin=1.1.0=hd590300_1
  - brotli-python=1.1.0=py310hc6cd4ac_1
  - bson=0.5.9=py_0
  - bzip2=1.0.8=hd590300_5
  - c-ares=1.28.1=hd590300_0
  - c-blosc2=2.14.4=hb4ffafa_1
  - ca-certificates=2024.8.30=hbcca054_0
  - cached-property=1.5.2=hd8ed1ab_1
  - cached_property=1.5.2=pyha770c72_1
  - cachetools=5.3.3=pyhd8ed1ab_0
  - cairo=1.18.0=h3faef2a_0
  - certifi=2024.8.30=pyhd8ed1ab_0
  - cffi=1.16.0=py310h2fee648_0
  - charset-normalizer=3.3.2=pyhd8ed1ab_0
  - colorama=0.4.6=pyhd8ed1ab_0
  - comm=0.2.2=pyhd8ed1ab_0
  - conda=24.5.0=py310h06a4308_0
  - conda-libmamba-solver=24.1.0=pyhd8ed1ab_0
  - conda-package-handling=2.2.0=pyh38be061_0
  - conda-package-streaming=0.9.0=pyhd8ed1ab_0
  - contourpy=1.2.1=py310hd41b1e2_0
  - cuda-cudart=12.1.105=0
  - cuda-cupti=12.1.105=0
  - cuda-libraries=12.1.0=0
  - cuda-nvrtc=12.1.105=0
  - cuda-nvtx=12.1.105=0
  - cuda-opencl=12.4.127=0
  - cuda-runtime=12.1.0=0
  - cudatoolkit=11.8.0=h4ba93d1_13
  - cycler=0.12.1=pyhd8ed1ab_0
  - cython=3.0.10=py310hc6cd4ac_0
  - debugpy=1.8.1=py310hc6cd4ac_0
  - decorator=5.1.1=pyhd8ed1ab_0
  - defusedxml=0.7.1=pyhd8ed1ab_0
  - distro=1.9.0=pyhd8ed1ab_0
  - entrypoints=0.4=pyhd8ed1ab_0
  - exceptiongroup=1.2.0=pyhd8ed1ab_2
  - executing=2.0.1=pyhd8ed1ab_0
  - expat=2.6.2=h59595ed_0
  - ffmpeg=4.3=hf484d3e_0
  - fftw=3.3.10=nompi_hc118613_108
  - filelock=3.13.4=pyhd8ed1ab_0
  - fmt=10.2.1=h00ab1b0_0
  - font-ttf-dejavu-sans-mono=2.37=hab24e00_0
  - font-ttf-inconsolata=3.000=h77eed37_0
  - font-ttf-source-code-pro=2.038=h77eed37_0
  - font-ttf-ubuntu=0.83=h77eed37_1
  - fontconfig=2.14.2=h14ed4e7_0
  - fonts-conda-ecosystem=1=0
  - fonts-conda-forge=1=0
  - fonttools=4.51.0=py310h2372a71_0
  - fqdn=1.5.1=pyhd8ed1ab_0
  - freetype=2.12.1=h267a509_2
  - frozendict=2.4.2=py310h5eee18b_0
  - frozenlist=1.4.1=py310h2372a71_0
  - fsspec=2024.3.1=pyhca7485f_0
  - gmp=6.3.0=h59595ed_1
  - gmpy2=2.1.2=py310h3ec546c_1
  - gnutls=3.6.13=h85f3911_1
  - greenlet=3.0.3=py310hc6cd4ac_0
  - h11=0.14.0=pyhd8ed1ab_0
  - h2=4.1.0=pyhd8ed1ab_0
  - hdf4=4.2.15=h9772cbc_5
  - hdf5=1.14.3=nompi_h4f84152_100
  - hpack=4.0.0=pyh9f0ad1d_0
  - httpcore=1.0.5=pyhd8ed1ab_0
  - httpx=0.27.0=pyhd8ed1ab_0
  - hyperframe=6.0.1=pyhd8ed1ab_0
  - icu=73.2=h59595ed_0
  - idna=3.6=pyhd8ed1ab_0
  - importlib-metadata=7.1.0=pyha770c72_0
  - importlib_metadata=7.1.0=hd8ed1ab_0
  - importlib_resources=6.4.0=pyhd8ed1ab_0
  - ipykernel=6.29.3=pyhd33586a_0
  - ipython=8.22.2=pyh707e725_0
  - ipywidgets=8.1.2=pyhd8ed1ab_0
  - isoduration=20.11.0=pyhd8ed1ab_0
  - jedi=0.19.1=pyhd8ed1ab_0
  - jinja2=3.1.3=pyhd8ed1ab_0
  - joblib=1.4.0=pyhd8ed1ab_0
  - jpeg=9e=h166bdaf_2
  - json5=0.9.24=pyhd8ed1ab_0
  - jsonpatch=1.33=pyhd8ed1ab_0
  - jsonpointer=2.4=py310hff52083_3
  - jsonschema=4.21.1=pyhd8ed1ab_0
  - jsonschema-specifications=2023.12.1=pyhd8ed1ab_0
  - jsonschema-with-format-nongpl=4.21.1=pyhd8ed1ab_0
  - jupyter-lsp=2.2.5=pyhd8ed1ab_0
  - jupyter_client=8.6.1=pyhd8ed1ab_0
  - jupyter_core=5.7.2=py310hff52083_0
  - jupyter_events=0.10.0=pyhd8ed1ab_0
  - jupyter_server=2.14.0=pyhd8ed1ab_0
  - jupyter_server_terminals=0.5.3=pyhd8ed1ab_0
  - jupyterlab=4.1.6=pyhd8ed1ab_0
  - jupyterlab_pygments=0.3.0=pyhd8ed1ab_1
  - jupyterlab_server=2.26.0=pyhd8ed1ab_0
  - jupyterlab_widgets=3.0.10=pyhd8ed1ab_0
  - keyutils=1.6.1=h166bdaf_0
  - kiwisolver=1.4.5=py310hd41b1e2_1
  - krb5=1.21.2=h659d440_0
  - lame=3.100=h166bdaf_1003
  - lcms2=2.12=hddcbb42_0
  - ld_impl_linux-64=2.40=h41732ed_0
  - lerc=3.0=h9c3ff4c_0
  - libaec=1.1.3=h59595ed_0
  - libarchive=3.7.2=h2aa1ff5_1
  - libblas=3.9.0=16_linux64_mkl
  - libbrotlicommon=1.1.0=hd590300_1
  - libbrotlidec=1.1.0=hd590300_1
  - libbrotlienc=1.1.0=hd590300_1
  - libcblas=3.9.0=16_linux64_mkl
  - libcublas=12.1.0.26=0
  - libcufft=11.0.2.4=0
  - libcufile=1.9.1.3=0
  - libcurand=10.3.5.147=0
  - libcurl=8.7.1=hca28451_0
  - libcusolver=11.4.4.55=0
  - libcusparse=12.0.2.55=0
  - libdeflate=1.10=h7f98852_0
  - libedit=3.1.20191231=he28a2e2_2
  - libev=4.33=hd590300_2
  - libexpat=2.6.2=h59595ed_0
  - libffi=3.4.2=h7f98852_5
  - libgcc=14.2.0=h77fa898_1
  - libgcc-ng=14.2.0=h69a702a_1
  - libgfortran-ng=13.2.0=h69a702a_5
  - libgfortran5=13.2.0=ha4646dd_5
  - libglib=2.80.0=hf2295e7_4
  - libgomp=14.2.0=h77fa898_1
  - libhwloc=2.9.3=default_h554bfaf_1009
  - libiconv=1.17=hd590300_2
  - libjpeg-turbo=2.0.0=h9bf148f_0
  - liblapack=3.9.0=16_linux64_mkl
  - liblapacke=3.9.0=16_linux64_mkl
  - libmamba=1.5.8=had39da4_0
  - libmambapy=1.5.8=py310h39ff949_0
  - libnetcdf=4.9.2=nompi_h9612171_113
  - libnghttp2=1.58.0=h47da74e_1
  - libnpp=12.0.2.50=0
  - libnsl=2.0.1=hd590300_0
  - libnvjitlink=12.1.105=0
  - libnvjpeg=12.1.1.14=0
  - libpng=1.6.43=h2797004_0
  - libsodium=1.0.18=h36c2ea0_1
  - libsolv=0.7.28=hfc55251_2
  - libsqlite=3.45.2=h2797004_0
  - libssh2=1.11.0=h0841786_0
  - libstdcxx-ng=13.2.0=h7e041cc_5
  - libtiff=4.3.0=h0fcbabc_4
  - libuuid=2.38.1=h0b41bf4_0
  - libwebp-base=1.3.2=hd590300_1
  - libxcb=1.15=h0b41bf4_0
  - libxcrypt=4.4.36=hd590300_1
  - libxml2=2.12.6=h232c23b_1
  - libxslt=1.1.39=h76b75d6_0
  - libzip=1.10.1=h2629f0a_3
  - libzlib=1.2.13=hd590300_5
  - lightning-utilities=0.11.2=pyhd8ed1ab_0
  - llvm-openmp=15.0.7=h0cdce71_0
  - loguru=0.7.2=py310hff52083_2
  - lxml=5.1.0=py310hcfd0673_0
  - lz4-c=1.9.4=hcb278e6_0
  - lzo=2.10=h516909a_1000
  - mamba=1.5.8=py310h51d5547_0
  - markupsafe=2.1.5=py310h2372a71_0
  - matplotlib-base=3.8.4=py310h62c0568_0
  - matplotlib-inline=0.1.6=pyhd8ed1ab_0
  - mda-xdrlib=0.2.0=pyhd8ed1ab_0
  - mdtraj=1.9.9=py310h523e8d7_1
  - menuinst=2.0.2=py310hff52083_0
  - mistune=3.0.2=pyhd8ed1ab_0
  - mkl=2022.1.0=h84fe81f_915
  - mkl-devel=2022.1.0=ha770c72_916
  - mkl-include=2022.1.0=h84fe81f_915
  - mpc=1.3.1=hfe3b2da_0
  - mpfr=4.2.1=h9458935_1
  - mpmath=1.3.0=pyhd8ed1ab_0
  - multidict=6.0.5=py310h2372a71_0
  - munkres=1.1.4=pyh9f0ad1d_0
  - nbclient=0.10.0=pyhd8ed1ab_0
  - nbconvert-core=7.16.3=pyhd8ed1ab_1
  - nbformat=5.10.4=pyhd8ed1ab_0
  - ncurses=6.4.20240210=h59595ed_0
  - nest-asyncio=1.6.0=pyhd8ed1ab_0
  - netcdf-fortran=4.6.1=nompi_hacb5139_103
  - nettle=3.6=he412f7d_0
  - networkx=3.2=pyhd8ed1ab_0
  - notebook=7.1.2=pyhd8ed1ab_0
  - notebook-shim=0.2.4=pyhd8ed1ab_0
  - ocl-icd=2.3.2=hd590300_1
  - ocl-icd-system=1.0.0=1
  - olefile=0.47=pyhd8ed1ab_0
  - openbabel=3.1.1=py310hbff9852_9
  - openff-amber-ff-ports=0.0.4=pyhca7485f_0
  - openff-forcefields=2024.03.0=pyhca7485f_0
  - openff-interchange=0.3.25=pyhd8ed1ab_1
  - openff-interchange-base=0.3.25=pyhd8ed1ab_1
  - openff-models=0.1.2=pyhca7485f_0
  - openff-toolkit=0.15.2=pyhd8ed1ab_0
  - openff-toolkit-base=0.15.2=pyhd8ed1ab_0
  - openff-units=0.2.2=pyhca7485f_0
  - openff-utilities=0.1.12=pyhd8ed1ab_0
  - openh264=2.1.1=h780b84a_0
  - openjpeg=2.5.0=h7d73246_0
  - openmm=8.1.1=py310h43b6314_1
  - openmmforcefields=0.12.0=pyhd8ed1ab_0
  - openssl=3.4.0=hb9d3cd8_0
  - overrides=7.7.0=pyhd8ed1ab_0
  - packaging=24.0=pyhd8ed1ab_0
  - packmol=20.010=h86c2bf4_0
  - pandas=2.1.2=py310hcc13569_0
  - pandocfilters=1.5.0=pyhd8ed1ab_0
  - panedr=0.8.0=pyhd8ed1ab_0
  - parmed=4.2.2=py310hc6cd4ac_1
  - parso=0.8.4=pyhd8ed1ab_0
  - pcre2=10.43=hcad00b1_0
  - pdbfixer=1.9=pyh1a96a4e_0
  - perl=5.32.1=7_hd590300_perl5
  - pexpect=4.9.0=pyhd8ed1ab_0
  - pickleshare=0.7.5=py_1003
  - pillow=8.4.0=py310h07f4688_0
  - pint=0.23=pyhd8ed1ab_0
  - pip=24.0=pyhd8ed1ab_0
  - pixman=0.43.2=h59595ed_0
  - pkgutil-resolve-name=1.3.10=pyhd8ed1ab_1
  - platformdirs=4.2.0=pyhd8ed1ab_0
  - pluggy=1.4.0=pyhd8ed1ab_0
  - plyfile=1.0.1=pyhd8ed1ab_0
  - prody=2.4.0=py310heca2aa9_0
  - prometheus_client=0.20.0=pyhd8ed1ab_0
  - prompt-toolkit=3.0.42=pyha770c72_0
  - psutil=5.9.8=py310h2372a71_0
  - pthread-stubs=0.4=h36c2ea0_1001
  - ptyprocess=0.7.0=pyhd3deb0d_0
  - pure_eval=0.2.2=pyhd8ed1ab_0
  - py-cpuinfo=9.0.0=pyhd8ed1ab_0
  - pybind11-abi=4=hd8ed1ab_3
  - pycairo=1.26.0=py310hda9f760_0
  - pycosat=0.6.6=py310h2372a71_0
  - pycparser=2.22=pyhd8ed1ab_0
  - pydantic=2.7.0=pyhd8ed1ab_0
  - pydantic-core=2.18.1=py310hcb5633a_0
  - pyedr=0.8.0=pyhd8ed1ab_0
  - pyg=2.5.2=py310_torch_2.2.0_cu121
  - pygments=2.17.2=pyhd8ed1ab_0
  - pyparsing=3.1.2=pyhd8ed1ab_0
  - pysocks=1.7.1=pyha2e5f31_6
  - pytables=3.9.2=py310hd76cd5d_2
  - python=3.10.0=h543edf9_3_cpython
  - python-constraint=1.4.0=py_0
  - python-dateutil=2.9.0=pyhd8ed1ab_0
  - python-fastjsonschema=2.19.1=pyhd8ed1ab_0
  - python-json-logger=2.0.7=pyhd8ed1ab_0
  - python-tzdata=2024.1=pyhd8ed1ab_0
  - python_abi=3.10=4_cp310
  - pytorch=2.2.2=py3.10_cuda12.1_cudnn8.9.2_0
  - pytorch-cuda=12.1=ha16c6d3_5
  - pytorch-mutex=1.0=cuda
  - pytz=2024.1=pyhd8ed1ab_0
  - pyyaml=6.0.1=py310h2372a71_1
  - pyzmq=25.1.2=py310h795f18f_0
  - readline=8.2=h8228510_1
  - reduce=3.24=0
  - referencing=0.34.0=pyhd8ed1ab_0
  - reportlab=3.5.68=py310h94fcab3_1
  - reproc=14.2.4.post0=hd590300_1
  - reproc-cpp=14.2.4.post0=h59595ed_1
  - requests=2.31.0=pyhd8ed1ab_0
  - rfc3339-validator=0.1.4=pyhd8ed1ab_0
  - rfc3986-validator=0.1.1=pyh9f0ad1d_0
  - rpds-py=0.18.0=py310hcb5633a_0
  - ruamel.yaml=0.18.6=py310h2372a71_0
  - ruamel.yaml.clib=0.2.8=py310h2372a71_0
  - scipy=1.8.1=py310hdfbd76f_2
  - send2trash=1.8.3=pyh0d859eb_0
  - setuptools=69.2.0=pyhd8ed1ab_0
  - six=1.16.0=pyh6c4a22f_0
  - smirnoff99frosst=1.1.0=pyh44b312d_0
  - snappy=1.2.0=hdb0a2a9_1
  - sniffio=1.3.1=pyhd8ed1ab_0
  - soupsieve=2.5=pyhd8ed1ab_1
  - sqlalchemy=2.0.29=py310h2372a71_0
  - sqlite=3.45.2=h2c6b66d_0
  - stack_data=0.6.2=pyhd8ed1ab_0
  - sympy=1.12=pypyh9d50eac_103
  - tbb=2021.11.0=h00ab1b0_1
  - terminado=0.18.1=pyh0d859eb_0
  - threadpoolctl=3.4.0=pyhc1e730c_0
  - tinycss2=1.2.1=pyhd8ed1ab_0
  - tinydb=4.8.0=pyhd8ed1ab_0
  - tk=8.6.13=noxft_h4845f30_101
  - tomli=2.0.1=pyhd8ed1ab_0
  - torch-ema=0.3=pyhd8ed1ab_0
  - torchaudio=2.2.2=py310_cu121
  - torchmetrics=1.2.1=pyhd8ed1ab_0
  - torchtriton=2.2.0=py310
  - torchvision=0.17.2=py310_cu121
  - tornado=6.4=py310h2372a71_0
  - tqdm=4.66.2=pyhd8ed1ab_0
  - traitlets=5.14.2=pyhd8ed1ab_0
  - truststore=0.8.0=pyhd8ed1ab_0
  - types-python-dateutil=2.9.0.20240316=pyhd8ed1ab_0
  - typing-extensions=4.11.0=hd8ed1ab_0
  - typing_extensions=4.11.0=pyha770c72_0
  - typing_utils=0.1.0=pyhd8ed1ab_0
  - tzdata=2024a=h0c530f3_0
  - unicodedata2=15.1.0=py310h2372a71_0
  - uri-template=1.3.0=pyhd8ed1ab_0
  - urllib3=2.2.1=pyhd8ed1ab_0
  - validators=0.28.0=pyhd8ed1ab_0
  - wcwidth=0.2.13=pyhd8ed1ab_0
  - webcolors=1.13=pyhd8ed1ab_0
  - webencodings=0.5.1=pyhd8ed1ab_2
  - websocket-client=1.7.0=pyhd8ed1ab_0
  - wheel=0.43.0=pyhd8ed1ab_1
  - widgetsnbextension=4.0.10=pyhd8ed1ab_0
  - xmltodict=0.13.0=pyhd8ed1ab_0
  - xorg-kbproto=1.0.7=h7f98852_1002
  - xorg-libice=1.1.1=hd590300_0
  - xorg-libsm=1.2.4=h7391055_0
  - xorg-libx11=1.8.9=h8ee46fc_0
  - xorg-libxau=1.0.11=hd590300_0
  - xorg-libxdmcp=1.1.3=h7f98852_0
  - xorg-libxext=1.3.4=h0b41bf4_2
  - xorg-libxrender=0.9.11=hd590300_0
  - xorg-libxt=1.3.0=hd590300_1
  - xorg-renderproto=0.11.1=h7f98852_1002
  - xorg-xextproto=7.3.0=h0b41bf4_1003
  - xorg-xproto=7.0.31=h7f98852_1007
  - xz=5.2.6=h166bdaf_0
  - yaml=0.2.5=h7f98852_2
  - yaml-cpp=0.8.0=h59595ed_0
  - yarl=1.9.4=py310h2372a71_0
  - zeromq=4.3.5=h59595ed_1
  - zipp=3.17.0=pyhd8ed1ab_0
  - zlib=1.2.13=hd590300_5
  - zlib-ng=2.0.7=h0b41bf4_0
  - zstandard=0.22.0=py310h1275a96_0
  - zstd=1.5.5=hfc55251_0
  - pip:
      - --find-links https://data.pyg.org/whl/torch-2.2.2%2Bcu121.html
      - accelerate==0.15.0
      - amberlite==22.0
      - amberutils==21.0
      - appdirs==1.4.4
      - biopython==1.79
      - click==8.1.7
      - dimorphite-dl==2.0.2
      - docker-pycreds==0.4.0
      - e3nn==0.5.1
      - et-xmlfile==1.1.0
      - fair-esm==2.0.0
      - fasteners==0.19
      - gitdb==4.0.11
      - gitpython==3.1.43
      - griddataformats==1.0.2
      - gsd==3.2.1
      - huggingface-hub==0.17.3
      - jupyter==1.0.0
      - jupyter-console==6.6.3
      - mdanalysis==2.4.0
      - mmpbsa-py==16.0
      - mmtf-python==1.1.3
      - mrcfile==1.5.0
      - msgpack==1.0.8
      - numexpr==2.10.0
      - numpy==1.24.4
      - openpyxl==3.1.3
      - opt-einsum==3.3.0
      - opt-einsum-fx==0.1.4
      - packmol-memgen==1.2.3rc0
      - pdb4amber==22.0
      - posebusters==0.2.13
      - prefetch-generator==1.0.3
      - protobuf==4.25.3
      - pyg-lib==0.4.0+pt22cu121 
      - pymesh==1.0.2
      - https://github.com/nuvolos-cloud/PyMesh/releases/download/v0.3.1/pymesh2-0.3.1-cp310-cp310-linux_x86_64.whl
      - pympler==1.0.1
      - pytraj==2.0.6
      - qtconsole==5.5.2
      - qtpy==2.4.1
      - rdkit==2023.3.1
      - rdkit-pypi==2022.9.5
      - regex==2023.12.25
      - sander==22.0
      - scikit-learn==1.3.2
      - sentry-sdk==1.45.0
      - setproctitle==1.3.3
      - smmap==5.0.1
      - spyrmsd==0.7.0
      - tokenizers==0.13.3
      - torch-cluster==1.6.3+pt22cu121
      - torch-scatter==2.1.2+pt22cu121
      - torch-sparse==0.6.18+pt22cu121
      - torch-spline-conv==1.2.2+pt22cu121
      - transformers==4.29.2
      - wandb==0.16.1

[Feriolet]

I tried to run SurfDock using the above environment, but unfortunately it does not work. I have found that the pymesh requires GLIBC 2.36, as mentioned in one of their issues Intelligent-Drug-Discovery-Lab/SurfDock#22. Im not sure why this pymesh requires an updated version of their ldd this time, where previously they require 2.34.

This would either mean we have to update the Ubuntu version to 24.04, or perhaps try to build pymesh from source

[DrrDom]

That is why a docker container may be a better solution in cases where versions of core libraries are mismatched.

My experiments with Boltz2 showed that it ignores protonation and tautomeric states. Thus, hydrogens added after "docking" to PDB structure of a ligand may look irrelevant. It also ignores input stereoconfiguration and generates an own configuration, thus docking of a specific stereoisomer may be impossible. In general it feels like it depends on applicability domain of the model, because for a large set of compounds it returns low predicted probabilities. However, this is difficult to truly estimate without prospective validation. The predicted probabilities poorly correlate with predictions of vina and gnina.

In general integration of Boltz2 may be challenging due to all these issues. SurfDock may have the same or similar issues.

[Feriolet]

Do boltz2 happen to remove all hydrogens in your input? Im not sure why but it feels many AI/ML models strips hydrogens away during training protein/ligand models in my experience.

[DrrDom]

Yes, it seems so, because no hydrogens is in output even if I supply explicit Hs in input SMILES. The developers mentioned that they use RemoveHs option and the maximum recommended size of a ligand is 56 atoms - https://github.com/jwohlwend/boltz/blob/main/docs/prediction.md#properties-affinity.

Removing of hydrogens may not be a problem if heavy atoms would be considered as augmented with hydrogens, like NH instead of N during the generation, but generation of different tautomers says that this is not the case.

[Feriolet]

They may have adjusted the tautomers then. I tried to find any file resembling tautomer processing, but the closest modification I could find is that they constraint the stereochemistry for the ligandd. I am not sure if this is the case, since it is difficult for me to understand how they processed each ligand (https://github.com/jwohlwend/boltz/blob/main/src/boltz/data/parse/schema.py).

For surfdock, I am able to mostly run their script with ubuntu24.04, but got stuck in their inference step, as I do not have any GPU in my laptop. They processed around 35 ligands in 5 hours, which is incredibly slow. I may try compiling the dockerfile again with torch cpu to see if the cpu version can optmisie the inference.

[Feriolet]

For the meantime, here is the setup I used to test surfdock. The yaml and dockerfile did not include installation of esm model and unzipping the pdb2pqr and apbs file.

new_environment.yaml
dockerfile copy.txt

Specs:
Mac OS M1 (8 CPU, 0 GPU)
14 GB of RAM is used to build docker
33 GB of storage was used to build docker

Docker specs:
Ubuntu 22.04, AMD
CUDA 12.8.0
python3.10

Time to dock 100 ligands in SurfDock's screen_pipeline.sh: 58652.98147010803 (16.3 hrs)

I also was not able to utilise multiprocessing for docking, so that is one of the reason why it takes so much time, especially since 1 molecule has 40 conformers. I think with GPU, it should take lesser time.

[DrrDom]

I created a container and converted it to apptainer image as we have only it on a cluster. After some tweaks screen_pipeline.sh started to work. However, there are some strange errors.

Why esm was not included into installation? Is it enough to clone it to the directory of SurfDock? I do not see a reason to use pip install -e . as suggested by developers, because it will not be installed in a conda environment inside a container. How did you install esm?

[Feriolet]

I installed fair-esm instead of the esm, as can be seen in the new_environment.yaml file. Indeed, I just git clone the esm repo so I can have access with the esm model. The environment.yaml in the Surfdock repository also did not have esm there, so I am not sure if it is necessary.