Merge pull request #21 from pangenome/Dockerfile

Dockerfile

.github/workflows/build_and_test_docker.yml

@@ -9,5 +9,7 @@ jobs:
       - uses: actions/checkout@v2
       - name: Build the Docker image
         run: docker build . --file Dockerfile --tag pgge
-      - name: Run a test on the DRB1-3123 dataset
-        run: docker run -v ${PWD}/data/:/data pgge "pgge -g "/data/HLA/DRB1-3123/*.consensus*.gfa" -f /data/HLA/DRB1-3123/DRB1-3123.fa -r /scripts/beehave.R -t 2 -o /data/HLA/DRB1-3123/pgge_docker -l 1000 -s 1000 -p 100"
+      - name: Run a test on the DRB1-3123 dataset dbg
+        run: docker run -v ${PWD}/data/:/data pgge "pgge -g '/data/HLA/DRB1-3123/*.consensus*.gfa' -f /data/HLA/DRB1-3123/DRB1-3123.fa -r /scripts/beehave.R -t 2 -o /data/HLA/DRB1-3123/pgge_docker -l 1000 -s 1000 -p 100"
+      - name: Run a test on the DRB1-3123 dataset vg
+        run: docker run -v ${PWD}/data/:/data pgge "pgge -g '/data/HLA/DRB1-3123/*.consensus*.gfa' -f /data/HLA/DRB1-3123/DRB1-3123.fa -r /scripts/beehave.R -t 2 -o /data/HLA/DRB1-3123/pgge_docker -l 1000 -s 2000 -p 10 --graphaligner-vg-mode"

Dockerfile

@@ -8,15 +8,12 @@ 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 \
@@ -62,21 +59,6 @@ RUN wget \
 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
@@ -93,4 +75,4 @@ 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" ]
+ENTRYPOINT [ "/bin/sh", "-c", "-l" ]

README.md

@@ -159,8 +159,9 @@ 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"
+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"
 ```
+:warning: In contrast to running `pgge` from the command line, when running in a docker container, we have to use `'` instead of `"` in order to ensure that the regex is parsed properly.
 
 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}`.
 
@@ -173,7 +174,7 @@ 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"
+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

environment.yml

@@ -9,3 +9,6 @@ dependencies:
   - conda-forge::r-ggrepel=0.9.1
   - conda-forge::r-gridextra=2.3
   - conda-forge::r-tidyverse=1.3.0
+  - conda-forge::open-fonts=0.7.0
+  - bioconda::samtools=1.12
+  - bioconda::graphaligner=1.0.13

pgge

@@ -15,6 +15,8 @@ beehave_R=false
 peanut_bed=false
 subsample_percentage=false
 subsample_number=false
+alignment_mode="dbg"
+graphaligner_vg_mode=false
 threads=1
 
 num_gfas=0
@@ -29,7 +31,7 @@ fi
 
 # read the options
 cmd=$0" "$@
-TEMP=`getopt -o g:f:i:n:r:o:l:s:p:u:t:hb --long input-gfa:,input-fasta:,input-gaf:,input-graph-names:,beehave-r:,output-dir:,seq-length:,step:,subsample-percentage:,subsample-number:,threads:,help,peanut-bed -n 'pgge' -- "$@"`
+TEMP=`getopt -o g:f:i:n:r:o:l:s:p:u:t:hbv --long input-gfa:,input-fasta:,input-gaf:,input-graph-names:,beehave-r:,output-dir:,seq-length:,step:,subsample-percentage:,subsample-number:,threads:,help,peanut-bed,graphaligner-vg-mode -n 'pgge' -- "$@"`
 eval set -- "$TEMP"
 
 # extract options and their arguments into variables.
@@ -46,6 +48,7 @@ while true ; do
         -p|--subsample-percentage) subsample_percentage=$2 ; shift 2 ;;
         -u|--subsample-number) subsample_number=$2 ; shift 2 ;;
         -b|--peanut-bed) peanut_bed=true ; shift ;;
+        -v|--graphaligner-vg-mode) graphaligner_vg_mode=true ; shift ;;
         -t|--threads) threads=$2 ; shift 2 ;;
         -h|--help) show_help=true ; shift ;;
         --) shift ; break ;;
@@ -132,6 +135,11 @@ then
     prefix_pgge="$prefix_pgge"-p"$subsample_percentage"
 fi
 
+if [[ "$graphaligner_vg_mode" != false ]];
+then    
+    alignment_mode="vg"
+fi
+
 if [ $show_help ];
 then
     #padding=`printf %${#0}s` # prints as many spaces as the length of $0
@@ -147,6 +155,7 @@ then
     echo "    -r, --beehave-r PATH         path to beehave.R"
     echo "    -b, --peanut-bed PATH        output BED file"  
     echo "    -n, --input-graph-names TSV  input graph name file: first row is the name of the original input file, second row is the display name in the PNG"    
+    echo "    -v, --graphaligner-vg-mode   run GraphAligner with -x vg (default is -x dbg)"    
     echo "   [splitfa]"
     echo "    -l, --seq-length N           length of the splits"
     echo "    -s, --step N                 step size between splits"
@@ -334,7 +343,7 @@ do
             -g "$gfa" \
             -f "$graph_aligner_fasta_input" \
             -a "$prefix_pgge"."$n"."$gfa_base".gaf \
-            -x vg \
+            -x "$alignment_mode" \
             -t "$threads" \
             2> >(tee -a "$log_file")
         ("$timer" -f "$fmt" cut -f 2,3,4,16 "$prefix_pgge"."$n"."$gfa_base".gaf \