Dockerfile, CI

Dockerfile

@@ -0,0 +1,95 @@
+FROM debian:bullseye-slim
+
+LABEL authors="Simon Heumos, Jean Monlong"
+LABEL description="Preliminary docker image containing all requirements for pgge pipeline"
+LABEL base_image="debian:bullseye-slim"
+LABEL software="pgge"
+LABEL about.home="https://github.com/pangenome/pgge"
+LABEL about.license="SPDX:MIT"
+
+# Required dependencies
+# samtools
+# TODO add samtools from Bioconda?
+RUN apt-get update \
+        && apt-get install -y --no-install-recommends \
+        wget \
+        curl \
+        less \
+        gcc \ 
+        samtools \
+        tzdata \
+        make \
+        git \
+        sudo \
+        pkg-config \
+        bc \
+        time \
+        libxml2-dev libssl-dev libcurl4-openssl-dev \ 
+        apt-transport-https software-properties-common dirmngr gpg-agent \ 
+        && rm -rf /var/lib/apt/lists/*
+
+# rust
+RUN curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs -o rust.sh && \
+	sh rust.sh -y --no-modify-path
+
+ENV PATH /root/.cargo/bin:$PATH
+
+RUN cd ../../
+# peanut
+RUN git clone https://github.com/pangenome/rs-peanut.git \
+    && cd rs-peanut \
+	&& git pull \
+    && git checkout 2783bca \
+	&& cargo build --release \
+    && cp target/release/peanut /usr/local/bin/peanut \
+    && cd ../
+
+# splitfa
+RUN git clone https://github.com/ekg/splitfa.git \
+    && cd splitfa \
+    && git pull \
+    && git checkout 98589b2 \
+	&& cargo build --release \
+    && cp target/release/splitfa /usr/local/bin/splitfa \
+    && cd ../
+
+# miniconda3
+RUN wget \
+    https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh \
+    && bash Miniconda3-latest-Linux-x86_64.sh -b -p /miniconda \
+    && rm -f Miniconda3-latest-Linux-x86_64.sh 
+ENV PATH /miniconda/bin:$PATH
+SHELL ["/bin/bash", "-c"]
+
+# GraphAligner
+# Unfortunately, the current Bioconda version of GraphAligner emits a 15-column GAF, whereas the most recent commit on github emits a 16-column GAF
+# Therefore, we can't use the Bioconda version as of now
+RUN git clone --recursive https://github.com/maickrau/GraphAligner \
+        && cd GraphAligner \
+        && git pull \
+        && git checkout 48143da \
+        && git submodule update --init --recursive \
+        && conda env create -f CondaEnvironment.yml \ 
+        && source activate GraphAligner \
+        && make bin/GraphAligner \
+        && cp bin/GraphAligner /usr/local/bin/GraphAligner \
+        && cd ../ \
+        && exit
+
+# Install the conda environment
+COPY environment.yml /
+RUN conda env create --quiet -f /environment.yml && conda clean -a
+
+# Add conda installation dir to PATH (instead of doing 'conda activate')
+ENV PATH /miniconda/envs/pgge-dev/bin:$PATH
+
+# Set path for all users
+RUN echo "export PATH=$PATH" > /etc/profile
+
+# bring in the binaries and scripts from pgge
+COPY pgge /usr/local/bin/pgge
+RUN mkdir /scripts
+COPY scripts/beehave.R /scripts/beehave.R
+RUN chmod 777 /usr/local/bin/pgge && chmod 777 /scripts/beehave.R
+
+ENTRYPOINT [ "/bin/bash", "-l", "-c" ]

README.md

@@ -135,6 +135,47 @@ _`pgge`_ also generates a visualization of the results `pgge_yeast/pgge-l100000-
 
 6. _[R](https://www.r-project.org/)_ with packages _[tidyverse](https://www.tidyverse.org/)_, _[ggrepel](https://www.rdocumentation.org/packages/ggrepel/versions/0.9.1)_, _[gridExtra](https://www.rdocumentation.org/packages/gridExtra/versions/2.3)_ installed.
 
+### docker
+
+To simplify installation and versioning, we have an automated GitHub action that pushes the current docker build to the GitHub registry.
+To use it, first pull the actual image:
+
+```sh
+docker pull ghcr.io/pangenome/pgge:latest
+```
+
+Or if you want to pull a specific snapshot from [https://github.com/orgs/pangenome/packages/container/package/pgge](https://github.com/orgs/pangenome/packages/container/package/pgge):
+
+```sh
+docker pull ghcr.io/pangenome/pgge:TAG
+```
+
+Going in the `pgge` directory
+
+```sh
+git clone --recursive https://github.com/pangenome/pgge.git
+cd pgge
+```
+
+you can run the container using the example [DRB1-3123](data/HLA/DRB1-3123) provided in this repo:
+```sh
+docker run -it -v ${PWD}/data/:/data pangenome/pgge "pgge -g "/data/HLA/DRB1-3123/*.consensus*.gfa" -f /data/HLA/DRB1-3123/DRB1-3123.fa -r /scripts/beehave.R -t 16 -o /data/HLA/DRB1-3123/pgge_docker -l 1000 -s 1000 -p 100"
+```
+
+The `-v` argument of `docker run` always expects a full path: `If you intended to pass a host directory, use absolute path.` This is taken care of by using `${PWD}`.
+
+If you want to experiment around, you can build a docker image locally using the `Dockerfile`:
+
+```sh
+docker build -t ${USER}/pgge:latest .
+```
+
+Staying in the `pgge` directory, we can run `pgge` with the locally build image:
+
+```sh
+docker run -it -v ${PWD}/data/:/data ${USER}/pgge "pgge -g "/data/HLA/DRB1-3123/*.consensus*.gfa" -f /data/HLA/DRB1-3123/DRB1-3123.fa -r /scripts/beehave.R -t 16 -o /data/HLA/DRB1-3123/pgge_docker -l 1000 -s 1000 -p 100"
+```
+
 ## TODOs
 - [x] _`pgge`_ should accept a list of GFA files as input (_path/to/files/\*.consensus\*.gfa_) and output the summarized results in one PNG
 - [x] Integrate https://github.com/ekg/splitfa as an option to prepare the input FASTA.
@@ -146,10 +187,10 @@ _`pgge`_ also generates a visualization of the results `pgge_yeast/pgge-l100000-
 - [ ] Add possibility to input several GAF files. Make sure the user can input a list of samples for the GAFs.
 - [ ] The user should be able to select options for GraphAligner.
 - [ ] Add a toolchain that compares the query alignments with the exact nodes they aligned to in the graph.
-- [ ] Add Dockerfile.
-- [ ] Add a CI building the Dockerfile and emitting evaluation metrics for all tools using `HLA-Zoo` data.
+- [x] Add Dockerfile.
+- [x] Add a CI building the Dockerfile and emitting evaluation metrics for all tools using `HLA-Zoo` data.
 - [ ] Add usage examples for _`minigraph`_, _`cactus`_, and _`SibeliaZ`_.
-- [ ] Integrate into nf-core/pangenome pipeline.
+- [ ] Integrate into nf-core/pangenome pipeline. `HALFWAY THERE`.
 
 ## authors