From 12e68b13bee8170fe088b6add5ca430cdb7a35d0 Mon Sep 17 00:00:00 2001
From: Asya Khleborodova <asyakhl@users.noreply.github.com>
Date: Tue, 19 Dec 2023 16:52:53 -0500
Subject: [PATCH] raven_assembler updated with docker image (#9)

---
 config/containers.json            |  1 +
 docker/raven_assembler/Dockerfile | 59 +++++++++++++++++++++++++++++++
 docker/raven_assembler/README.md  | 35 ++++++++++++++++++
 workflow/Snakefile                |  6 ++--
 4 files changed, 97 insertions(+), 4 deletions(-)
 create mode 100644 docker/raven_assembler/Dockerfile
 create mode 100644 docker/raven_assembler/README.md

diff --git a/config/containers.json b/config/containers.json
index 46aecee..49458a5 100644
--- a/config/containers.json
+++ b/config/containers.json
@@ -1,4 +1,5 @@
 {
     "images": {
+        "raven": "docker://asyakhleborodova/raven_assembly:v0.1.0"
     }
 }
diff --git a/docker/raven_assembler/Dockerfile b/docker/raven_assembler/Dockerfile
new file mode 100644
index 0000000..63e090f
--- /dev/null
+++ b/docker/raven_assembler/Dockerfile
@@ -0,0 +1,59 @@
+FROM ubuntu:22.04
+
+RUN mkdir -p /opt2 && mkdir -p /data2
+WORKDIR /opt2 
+
+# Set time zone to US east coast 
+ENV TZ=America/New_York
+RUN ln -snf /usr/share/zoneinfo/$TZ /etc/localtime \
+       && echo $TZ > /etc/timezone
+
+RUN apt-get update \
+   && apt-get -y upgrade \
+   && DEBIAN_FRONTEND=noninteractive apt-get install -y \
+       build-essential \
+       cmake \
+       curl \
+       gawk \
+       git \
+       gzip \
+       g++ \
+       libboost-dev \
+       libcurl4-openssl-dev \
+       libssl-dev \
+       libxml2-dev \
+       locales \
+       # python/3.10.6
+       python3 \
+       python3-pip \
+       samtools \
+       wget \
+       zlib1g-dev \
+   && apt-get clean && apt-get purge \
+   && rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
+
+
+ENV CONDA_DIR /opt2/conda
+RUN wget --quiet https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O ~/miniconda.sh && \
+    /bin/bash ~/miniconda.sh -b -p /opt2/conda
+
+# Put conda in path so we can use conda activate
+ENV PATH=$CONDA_DIR/bin:$PATH
+
+# Set the locale
+RUN localedef -i en_US -f UTF-8 en_US.UTF-8
+
+############### INSTALL ################
+RUN ln -sf /usr/bin/python3 /usr/bin/python
+RUN pip3 install --upgrade pip \
+   && pip3 install matplotlib \
+   && pip3 install numpy \
+   && pip3 install pandas \
+   && pip3 install scipy
+
+RUN conda install -y --channel=defaults --channel=bioconda --channel=conda-forge raven-assembler
+
+WORKDIR /opt2 
+
+COPY Dockerfile /opt2
+WORKDIR /data2
diff --git a/docker/raven_assembler/README.md b/docker/raven_assembler/README.md
new file mode 100644
index 0000000..454b395
--- /dev/null
+++ b/docker/raven_assembler/README.md
@@ -0,0 +1,35 @@
+## Steps for Building Docker Images
+
+Directly below are instructions for building an image using the provided Dockerfile:
+
+```bash
+# See listing of images on computer
+docker image ls
+
+# Build from Dockerfile
+docker build --no-cache -f Dockerfile --tag=raven_assembly:v0.1.0 .
+
+# Testing, take a peek inside
+docker run -ti raven_assembly:v0.1.0 /bin/bash
+
+# Updating Tag  before pushing to DockerHub
+docker tag raven_assembly:v0.1.0 asyakhleborodova/raven_assembly:v0.1.0
+docker tag raven_assembly:v0.1.0 asyakhleborodova/raven_assembly         # latest
+
+# Check out new tag(s)
+docker image ls
+
+# Push new tagged image to DockerHub
+docker push asyakhleborodova/raven_assembly:v0.1.0
+docker push asyakhleborodova/raven_assembly:latest
+```
+
+### Other Recommended Steps
+
+Scan your image for known vulnerabilities:
+
+```bash
+docker scan raven_assembly:v0.1.0
+```
+
+> **Please Note**: Any references to `asyakhleborodova` should be replaced your username if you would also like to push the image to a non-org account.
diff --git a/workflow/Snakefile b/workflow/Snakefile
index fefa0cc..df3c684 100644
--- a/workflow/Snakefile
+++ b/workflow/Snakefile
@@ -90,7 +90,7 @@ rule To_Fasta:
     params:
         rname="To_Fasta",
         samtools="samtools/1.9",
-        seqkit="seqkit/0.12.1",
+        seqkit="seqkit/2.2.0",
         dir=directory(join(workpath, "reads")),
         temp_ext = lambda w: ext[w.samples],
         FQ=join(workpath, "reads/{samples}.fastq")
@@ -130,17 +130,15 @@ rule raven_assembly:
         dir=directory(join(workpath, "raven_assembly")),
         gfa="{samples}.raven.gfa",
     #conda: "envs/raven-assembler.yaml"
+    container: config['images']['raven']
     threads: 32
     shell:
         """
-        source /data/NCBR/apps/genome-assembly/conda/etc/profile.d/conda.sh
-        conda activate raven-assembler
         mkdir -p {params.dir}
         cd {params.dir}
         #raven --threads {threads} {input} > {output.fa}
         raven --graphical-fragment-assembly {params.gfa} --threads {threads} {input}
         awk '$1 ~/S/ {{print ">"$2"\\n"$3}}' {output.gfa} > {output.fa}
-        conda deactivate
         """
 
 # rule wtdbg2_assembly: