Updated Dockerfile to address incompatibilities during setup

[mbassalbioinformatics]

This isn't an issue per-se but a suggestion and might be useful for others. In trying to setup the updated version of alphafold a bunch of incompatibilities came up when setting up the docker image, particularly around getting the right updated compatible versions of CUDA etc..

Took a while to get a version of it all that works but sharing it here in case it helps others. Im inexperienced with Docker so if anyone has a better way of setting up the Dockerfile to address these issues, then please by all means share!

# Copyright 2021 DeepMind Technologies Limited
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#      http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

# ARG CUDA=11.0
# FROM nvidia/cuda:${CUDA}-cudnn8-runtime-ubuntu18.04
ARG CUDA=11.4.1
FROM nvidia/cuda:${CUDA}-cudnn8-runtime-ubuntu20.04
# FROM directive resets ARGS, so we specify again (the value is retained if
# previously set).
ARG CUDA

# Use bash to support string substitution.
SHELL ["/bin/bash", "-c"]

RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y \
      build-essential \
      cmake \
      # cuda-command-line-tools-${CUDA/./-} \
      git \
      hmmer \
      kalign \
      tzdata \
      wget \
    && rm -rf /var/lib/apt/lists/*

CMD ["wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/cuda-ubuntu2004.pin"]
CMD ["sudo mv cuda-ubuntu2004.pin /etc/apt/preferences.d/cuda-repository-pin-600"]
# CMD ["wget https://developer.download.nvidia.com/compute/cuda/11.4.1/local_installers/cuda-repo-ubuntu2004-11-4-local_11.4.1-470.57.02-1_amd64.deb"]
CMD ["dpkg -i /app/alphafold/cuda-repo-ubuntu2004-11-4-local_11.4.1-470.57.02-1_amd64.deb"]
CMD ["apt-key add /var/cuda-repo-ubuntu2004-11-4-local/7fa2af80.pub"]
RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y cuda


# Compile HHsuite from source.
RUN git clone --branch v3.3.0 https://github.com/soedinglab/hh-suite.git /tmp/hh-suite \
    && mkdir /tmp/hh-suite/build \
    && pushd /tmp/hh-suite/build \
    && cmake -DCMAKE_INSTALL_PREFIX=/opt/hhsuite .. \
    && make -j 4 && make install \
    && ln -s /opt/hhsuite/bin/* /usr/bin \
    && popd \
    && rm -rf /tmp/hh-suite

# Install Miniconda package manager.
RUN wget -q -P /tmp \
  https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh \
    && bash /tmp/Miniconda3-latest-Linux-x86_64.sh -b -p /opt/conda \
    && rm /tmp/Miniconda3-latest-Linux-x86_64.sh

# Install conda packages.
ENV PATH="/opt/conda/bin:$PATH"
RUN conda update -qy conda \
    && conda install -y -c conda-forge \
      openmm=7.5.1 \
      # cudatoolkit==${CUDA_VERSION} \
      pdbfixer \
      pip \
      python=3.7

COPY . /app/alphafold
RUN wget -q -P /app/alphafold/alphafold/common/ \
  https://git.scicore.unibas.ch/schwede/openstructure/-/raw/7102c63615b64735c4941278d92b554ec94415f8/modules/mol/alg/src/stereo_chemical_props.txt

# Install pip packages.
RUN pip3 install --upgrade pip \
    && pip3 install -r /app/alphafold/requirements.txt \
    && pip3 install /app/alphafold/jaxlib-0.1.75+cuda11.cudnn805-cp37-none-manylinux2010_x86_64.whl
    # && pip3 install --upgrade jax jaxlib==0.1.69+cuda${CUDA/./} -f \
    #  https://storage.googleapis.com/jax-releases/jax_releases.html

# Apply OpenMM patch.
WORKDIR /opt/conda/lib/python3.7/site-packages
RUN patch -p0 < /app/alphafold/docker/openmm.patch

# We need to run `ldconfig` first to ensure GPUs are visible, due to some quirk
# with Debian. See https://github.com/NVIDIA/nvidia-docker/issues/1399 for
# details.
# ENTRYPOINT does not support easily running multiple commands, so instead we
# write a shell script to wrap them up.
WORKDIR /app/alphafold
RUN echo $'#!/bin/bash\n\
ldconfig\n\
python /app/alphafold/run_alphafold.py "$@"' > /app/run_alphafold.sh \
  && chmod +x /app/run_alphafold.sh
ENTRYPOINT ["/app/run_alphafold.sh"]

[Augustin-Zidek]

Thanks. I think this was fixed in https://github.com/deepmind/alphafold/releases/tag/v2.1.2, feel free to reopen this issue if that is not the case.

[mbassalbioinformatics]

Hi all

Setting up alphafold on a new machine and the same issues arose as last time. Below is the updated Dockerfile needed to be setup correctly and run. Hope its helpful for others.

# Copyright 2021 DeepMind Technologies Limited
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#      http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

ARG CUDA=12.1.0
# ARG CUDA=11.1.1
FROM nvidia/cuda:${CUDA}-runtime-ubuntu22.04
# FROM nvidia/cuda:${CUDA}-runtime
# FROM directive resets ARGS, so we specify again (the value is retained if
# previously set).
ARG CUDA
# ARG CUDA_VERSION=11.1.1

# Use bash to support string substitution.
SHELL ["/bin/bash", "-o", "pipefail", "-c"]

RUN apt-get update \
    && DEBIAN_FRONTEND=noninteractive apt-get install --no-install-recommends -y \
        build-essential \
        cmake \
        cuda-command-line-tools-$(cut -f1,2 -d- <<< ${CUDA//./-}) \
        git \
        hmmer \
        kalign \
        tzdata \
        wget \
    && rm -rf /var/lib/apt/lists/* \
    && apt-get autoremove -y \
    && apt-get clean

CMD ["wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-ubuntu2204.pin"]
CMD ["sudo mv cuda-ubuntu2204.pin /etc/apt/preferences.d/cuda-repository-pin-600"]
CMD ["wget https://developer.download.nvidia.com/compute/cuda/12.1.0/local_installers/cuda-repo-ubuntu2204-12-1-local_12.1.0-530.30.02-1_amd64.deb"]
CMD ["sudo dpkg -i cuda-repo-ubuntu2204-12-1-local_12.1.0-530.30.02-1_amd64.deb"]
CMD ["apt-key add /var/cuda-repo-ubuntu2204-12-1-local/3bf863cc.pub"]
RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y cuda


# Compile HHsuite from source.
RUN git clone --branch v3.3.0 https://github.com/soedinglab/hh-suite.git /tmp/hh-suite \
    && mkdir /tmp/hh-suite/build \
    && pushd /tmp/hh-suite/build \
    && cmake -DCMAKE_INSTALL_PREFIX=/opt/hhsuite .. \
    && make -j 4 && make install \
    && ln -s /opt/hhsuite/bin/* /usr/bin \
    && popd \
    && rm -rf /tmp/hh-suite

# Install Miniconda package manager.
RUN wget -q -P /tmp \
  https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh \
    && bash /tmp/Miniconda3-latest-Linux-x86_64.sh -b -p /opt/conda \
    && rm /tmp/Miniconda3-latest-Linux-x86_64.sh

# Install conda packages.
ENV PATH="/opt/conda/bin:$PATH"
# RUN conda install -qy conda==4.13.0 \
RUN conda install -qy conda=23.1.0 \
    && conda install -y -c conda-forge \
      openmm=7.5.1 \
      # cudatoolkit==${CUDA_VERSION} \
      pdbfixer \
      pip \
      python=3.8 \
      && conda clean --all --force-pkgs-dirs --yes

COPY . /app/alphafold
RUN wget -q -P /app/alphafold/alphafold/common/ \
  https://git.scicore.unibas.ch/schwede/openstructure/-/raw/7102c63615b64735c4941278d92b554ec94415f8/modules/mol/alg/src/stereo_chemical_props.txt

# Install pip packages.
RUN pip3 install --upgrade pip --no-cache-dir \
    && pip3 install -r /app/alphafold/requirements.txt --no-cache-dir \
    # && pip3 install --upgrade --no-cache-dir \
    && pip3 install https://storage.googleapis.com/jax-releases/cuda11/jaxlib-0.3.25+cuda11.cudnn82-cp38-cp38-manylinux2014_x86_64.whl
      # jax==0.3.25 \
      # jaxlib==0.3.25+cuda11.cudnn805 \
      # -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html

# Apply OpenMM patch.
WORKDIR /opt/conda/lib/python3.8/site-packages
RUN patch -p0 < /app/alphafold/docker/openmm.patch

# Add SETUID bit to the ldconfig binary so that non-root users can run it.
RUN chmod u+s /sbin/ldconfig.real

# We need to run `ldconfig` first to ensure GPUs are visible, due to some quirk
# with Debian. See https://github.com/NVIDIA/nvidia-docker/issues/1399 for
# details.
# ENTRYPOINT does not support easily running multiple commands, so instead we
# write a shell script to wrap them up.
WORKDIR /app/alphafold
RUN echo $'#!/bin/bash\n\
ldconfig\n\
python /app/alphafold/run_alphafold.py "$@"' > /app/run_alphafold.sh \
  && chmod +x /app/run_alphafold.sh
ENTRYPOINT ["/app/run_alphafold.sh"]