Merge pull request #3 from bacpop/docker

Docker

.github/workflows/docker_push.yml

@@ -0,0 +1,47 @@
+name: Build and push Docker image
+
+on:
+  push:
+    branches:
+      - 'main'
+      - 'docker'
+    tags:
+      - 'v*'
+  pull_request:
+    branches:
+      - 'main'
+  create:
+    tags:
+      - v*
+
+jobs:
+  docker-upload:
+    runs-on: ubuntu-latest
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v2
+    - name: Docker meta
+      id: meta
+      uses: docker/metadata-action@v4
+      with:
+        images: samhorsfield96/celebrimbor
+    - name: Set up QEMU
+      uses: docker/setup-qemu-action@v1
+    - name: Set up Docker Buildx
+      uses: docker/setup-buildx-action@v1
+    - name: Login to DockerHub
+      uses: docker/login-action@v1
+      with:
+        username: ${{ secrets.DOCKER_REGISTRY_USERNAME }}
+        password: ${{ secrets.DOCKER_REGISTRY_PASSWORD }}
+    - name: Build and push
+      id: docker_build
+      uses: docker/build-push-action@v3
+      with:
+        push: ${{ github.event_name != 'pull_request' }}
+        tags: ${{ steps.meta.outputs.tags }}
+        labels: ${{ steps.meta.outputs.labels }}
+        file: docker/Dockerfile
+        provenance: false
+    - name: Image digest
+      run: echo ${{ steps.docker_build.outputs.digest }}

README.md

@@ -38,9 +38,9 @@ You can also use the light bakta database if using a suitable version of bakta:
 bakta_db download --output /path/to/database --type light
 ```
 
-Install [cgt](https://github.com/bacpop/cgt)
+Install [cgt](https://github.com/bacpop/cgt) (will install `cgt_bacpop` executable in `./bin` directory)
 ```
-cargo add cgt_bacpop
+cargo install cgt_bacpop --root .
 ```
 
 Or to build from source:
@@ -50,6 +50,24 @@ cd cgt
 cargo install --path "."
 ```
 
+### Running inside a container
+
+An alternative, if you are having trouble with the above, is to use the CELEBRIMBOR docker
+container. If you are comfortable running commands inside docker containers and mounting
+your external files, the whole pipeline is in the container available by running:
+
+```
+docker pull samhorsfield96/celebrimbor:main
+```
+
+To run within the container, use the below command, replacing `path to output dir` and `path to fasta dir` with absolute paths and changing other parameters as required:
+
+```
+docker run -v <path to output dir>:/output -v <path to fasta dir>:/data samhorsfield96/celebrimbor:main snakemake --cores 4 --config genome_fasta=/data output_dir=/output bakta_db=bakta_db/db-light cgt_exe=cgt_bacpop cgt_breaks=0.05,0.95 cgt_error=0.05 clustering_method=panaroo panaroo_stringency=moderate
+```
+
+Note: ensure that `clustering_method` and `panaroo_stringency` parameters are not in quotes.
+
 ## Quick start:
 
 Update `config.yaml` to specify workflow and directory paths.

config.yaml

@@ -9,11 +9,11 @@ bakta_db: path/to/bakta/db/dir
 
 # cgt executable parameters
 cgt_exe: /path/to/cgt/exe
-cgt_breaks: 0.1,0.95
+cgt_breaks: 0.05,0.95
 cgt_error: 0.05
 
-# choice of clustering method, either 'mmseqs2' or 'panaroo'
+# choice of clustering method, either "mmseqs2" or "panaroo"
 clustering_method: "panaroo"
 
-# must be one of 'strict', 'moderate' or 'sensitive'
+# must be one of "strict", "moderate" or "sensitive"
 panaroo_stringency: "strict"

docker/.dockerignore

@@ -0,0 +1,3 @@
+../test/*
+test/*
+.git

docker/Dockerfile

@@ -0,0 +1,82 @@
+# From https://github.com/kaust-vislab/python-data-science-project
+FROM ubuntu:20.04
+
+LABEL maintainer="Sam Horsfield <[email protected]>"
+
+SHELL [ "/bin/bash", "--login", "-c" ]
+
+RUN apt-get update --fix-missing && \
+    apt-get install -y wget bzip2 curl git && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/*
+
+# Create a non-root user
+ARG username=celebrimbor-usr
+ARG uid=1000
+ARG gid=100
+ENV USER $username
+ENV UID $uid
+ENV GID $gid
+ENV HOME /home/$USER
+
+RUN adduser --disabled-password \
+    --gecos "Non-root user" \
+    --uid $UID \
+    --gid $GID \
+    --home $HOME \
+    $USER
+
+COPY environment.yml /tmp/
+RUN chown $UID:$GID /tmp/environment.yml
+
+COPY docker/entrypoint.sh /usr/local/bin/
+RUN chown $UID:$GID /usr/local/bin/entrypoint.sh && \
+    chmod u+x /usr/local/bin/entrypoint.sh
+
+USER $USER
+
+# install miniconda
+ENV MINICONDA_VERSION latest
+ENV CONDA_DIR $HOME/miniconda3
+RUN wget --quiet https://repo.anaconda.com/miniconda/Miniconda3-$MINICONDA_VERSION-Linux-x86_64.sh -O ~/miniconda.sh && \
+    chmod +x ~/miniconda.sh && \
+    ~/miniconda.sh -b -p $CONDA_DIR && \
+    rm ~/miniconda.sh
+
+# make non-activate conda commands available
+ENV PATH=$CONDA_DIR/bin:$PATH
+
+# make conda activate command available from /bin/bash --login shells
+RUN echo ". $CONDA_DIR/etc/profile.d/conda.sh" >> ~/.profile
+
+# make conda activate command available from /bin/bash --interative shells
+RUN conda init bash
+
+# create a project directory inside user home
+# (this isn't used with a clone running snakemake)
+ENV PROJECT_DIR $HOME/app
+RUN mkdir $PROJECT_DIR
+# copy the code in
+COPY . $PROJECT_DIR
+WORKDIR $PROJECT_DIR
+
+# build the conda environment
+ENV ENV_PREFIX $PROJECT_DIR/env
+RUN conda update --name base --channel defaults conda && \
+    conda env create --prefix $ENV_PREFIX --file /tmp/environment.yml --yes && \
+    conda clean --all --yes && \
+    conda activate $ENV_PREFIX && \
+    cargo install cgt_bacpop && \
+    mkdir bakta_db && \ 
+    bakta_db download --output bakta_db --type light && \
+    conda deactivate
+
+# add rust binary to path
+ENV PATH="$HOME/.cargo/bin:${PATH}"
+
+# use an entrypoint script to insure conda environment is properly activated at runtime
+USER root
+ENTRYPOINT [ "/usr/local/bin/entrypoint.sh" ]
+
+# default of running shell is fine
+#CMD [ "bash" ]

docker/entrypoint.sh

@@ -0,0 +1,5 @@
+#!/bin/bash --login
+set -e
+
+conda activate $HOME/app/env
+exec "$@"