Containerized StrVCTVRE

.dockerignore

@@ -0,0 +1,5 @@
+.git
+images
+.gitignore
+LICENSE
+data/hg38.phyloP100way.bw

Dockerfile

@@ -0,0 +1,102 @@
+# This Dockerfile builds a container image for StrVCTVRE, a tool for structural variant classification.
+# It uses micromamba to manage dependencies and sets up the environment for running StrVCTVRE.
+# It's also been built for use with Docker and Apptainer, so it can be run on HPC clusters.
+FROM mambaorg/micromamba:ubuntu24.04 AS micromamba
+
+FROM ubuntu:24.04
+
+RUN apt-get update \
+  && apt-get install -y --no-install-recommends \
+    wget \
+    curl \
+    ca-certificates \
+  && apt-get clean autoclean \
+  && apt-get autoremove -y \
+  && rm -rf /var/lib/apt/lists/* \
+  && rm -f /var/cache/apt/archives/*.deb
+
+#############################################################
+######################### General ###########################
+#############################################################
+RUN mkdir -p /software; chmod a+rwx -R /software; mkdir /software/strvctvre
+
+WORKDIR /software/strvctvre
+
+COPY . .
+
+# The console script (`StrVCTVRE = "StrVCTVRE:main"`) imports `main` from the
+# module, but the committed StrVCTVRE.py runs argparse + all logic at module
+# level and defines no `main()`. Wrap the entire script in `def main():` and add
+# a `__main__` guard, in-place in the image so the repo source stays untouched.
+RUN awk '\
+BEGIN { print "def main():" } \
+{ if ($0 ~ /^[[:space:]]*$/) print $0; else print "    " $0 } \
+END { print ""; print "if __name__ == \"__main__\":"; print "    main()" } \
+' /software/strvctvre/StrVCTVRE.py > /tmp/StrVCTVRE.py \
+  && mv /tmp/StrVCTVRE.py /software/strvctvre/StrVCTVRE.py
+
+# Update CERTs if building on VPN
+# RUN wget -P /tmp/ <cert_url>
+# ARG CURL_CA_BUNDLE=/tmp/<cert_file>
+
+# Install micromamba first
+# Adding micromamba to existing image:  https://micromamba-docker.readthedocs.io/en/latest/advanced_usage.html#adding-micromamba-to-an-existing-docker-image
+# https://github.com/mamba-org/micromamba-docker/blob/main/debian.Dockerfile
+
+# if your image defaults to a non-root user, then you may want to make the
+# next 3 ARG commands match the values in your image. You can get the values
+# by running: docker run --rm -it my/image id -a
+ARG MAMBA_USER=root
+ARG MAMBA_USER_ID=0
+ARG MAMBA_USER_GID=0
+ENV MAMBA_USER=$MAMBA_USER
+ENV MAMBA_ROOT_PREFIX="/software/conda"
+ENV MAMBA_EXE="/bin/micromamba"
+## conda env name (must match `name:` in environment_py3.yml)
+ENV ENV_NAME=StrVCTVRE_py_3
+#ENV MAMBA_SKIP_ACTIVATE=1; We want to activate, but just not the base
+
+COPY --from=micromamba "$MAMBA_EXE" "$MAMBA_EXE"
+COPY --from=micromamba /usr/local/bin/_activate_current_env.sh /usr/local/bin/_activate_current_env.sh
+COPY --from=micromamba /usr/local/bin/_dockerfile_shell.sh /usr/local/bin/_dockerfile_shell.sh
+COPY --from=micromamba /usr/local/bin/_entrypoint.sh /usr/local/bin/_entrypoint.sh
+COPY --from=micromamba /usr/local/bin/_dockerfile_initialize_user_accounts.sh /usr/local/bin/_dockerfile_initialize_user_accounts.sh
+COPY --from=micromamba /usr/local/bin/_dockerfile_setup_root_prefix.sh /usr/local/bin/_dockerfile_setup_root_prefix.sh
+
+RUN /usr/local/bin/_dockerfile_initialize_user_accounts.sh && \
+    /usr/local/bin/_dockerfile_setup_root_prefix.sh
+
+RUN --mount=type=cache,target=/software/conda/pkgs \
+    micromamba create -y -f /software/strvctvre/environment_py3.yml && \
+    micromamba clean --all --yes
+
+# Install StrVCTVRE in editable mode so that it's available as a console script (e.g. `StrVCTVRE --help` vs `python StrVCTVRE.py --help`)
+RUN --mount=type=cache,target=/software/conda/pkgs \
+    micromamba run -n $ENV_NAME pip --trusted-host pypi.org --trusted-host files.pythonhosted.org install -e .
+
+WORKDIR /software/
+#############################################################
+#############################################################
+#############################################################
+# Add the micromamba activation script to the system-wide bashrc so that the environment is activated for all users on run/exec for apptainer and docker.
+RUN echo "source /usr/local/bin/_activate_current_env.sh" >> /etc/bash.bashrc && \
+    echo "source /usr/local/bin/_activate_current_env.sh" >> .bash_env && \
+    echo "source /etc/bash.bashrc" >> .bash_env
+
+# any calls to bash will load this environment (e.g. docker/apptainer exec bash -c "cmd")
+ENV BASH_ENV=/software/.bash_env
+
+RUN chmod +x /software/strvctvre/strvctvre_entrypoint.sh
+
+ENV PATH=/software/strvctvre:/software:$PATH
+
+WORKDIR /software/strvctvre
+# The `pip install -e .` above installs the `StrVCTVRE` console script, but
+# setuptools flat-layout auto-discovery picks up the `data/` dir as the package
+# instead of the top-level StrVCTVRE.py module, so `from StrVCTVRE import main`
+# fails at runtime. Put the repo on PYTHONPATH so the module is importable.
+ENV PYTHONPATH=/software/strvctvre
+
+ENTRYPOINT ["/software/strvctvre/strvctvre_entrypoint.sh"]
+
+CMD ["/bin/bash"]

README.md

@@ -6,6 +6,36 @@ Structural variant impact predictor developed by Andrew Sharo, Zhiqiang Hu, and
 
 ### \*\*New\*\*: Visit our [Web Server](https://strvctvre.berkeley.edu) to annotate vcf or bed files with StrVCTVRE scores, or query a single SV
 
+## Run with Docker (optional)
+
+A `Dockerfile` is provided as an alternative to installing Python, conda, and the required packages manually (i.e. it replaces steps 1–4 below). From the cloned repository, build the image with:
+```bash
+git clone --depth 1 https://github.com/andrewSharo/StrVCTVRE
+cd StrVCTVRE
+docker build --platform linux/amd64 -t strvctvre:v1.10 .
+```
+This docker image creates a console script that can be run with the command `StrVCTVRE`.
+
+The `--platform linux/amd64` flag is required, including on Apple Silicon / ARM machines, where it builds under emulation.
+
+The image does **not** include the 9.2GB `hg38.phyloP100way.bw` file (see step 3). Download it separately and mount the folder containing it (and a folder for your input/output) into the container. For example, to annotate a vcf file:
+```bash
+cd data
+wget http://hgdownload.cse.ucsc.edu/goldenpath/hg38/phyloP100way/hg38.phyloP100way.bw
+cd ..
+
+# Try the test data
+docker run --rm --platform linux/amd64 \
+  -v $(pwd)/data:/software/strvctvre/data \
+  -v /path/to/your/files:/io \
+  strvctvre:v1.10 StrVCTVRE -i data/test.vcf.gz -o /io/output.vcf
+```
+
+## Run with apptainer/snakemake (optional)
+Snakemake runs apptainer exec mode.  By default apptainer will mount the current working directory and use it as in the container
+as the current working directory.  This means that you should have the data folder system linked in snakemake's working directory
+for the StrVCTVRE console script (or StrVCTVRE.py) to find the hg38.phyloP100way.bw file.
+
 ## To run StrVCTVRE, follow these steps:
 
 ### 1. Download and install Python 3 (if not done already)

pyproject.toml

@@ -0,0 +1,11 @@
+[build-system]
+requires = ["setuptools", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "StrVCTVRE"
+version = "0.1.0"
+dependencies = [] # your dependencies here
+
+[project.scripts]
+StrVCTVRE = "StrVCTVRE:main"

strvctvre_entrypoint.sh

@@ -0,0 +1,29 @@
+#!/bin/bash
+
+# StrVCTVRE container entrypoint.
+# Activates the conda environment, then warns if the required phyloP
+# conservation file is missing (it is ~9.2GB and is NOT baked into the image).
+
+IGreen='\033[0;92m'       # Green
+On_Red='\033[41m'         # Red
+Color_Off='\033[0m'       # Text Reset
+
+# Path StrVCTVRE.py checks by default; override at runtime with -p.
+PHYLOP_FILE="${PHYLOP_FILE:-data/hg38.phyloP100way.bw}"
+
+if [ ! -s "${PHYLOP_FILE}" ]; then
+  echo -e "${IGreen}The phyloP conservation file was not found at:\n  ${PHYLOP_FILE}${Color_Off}"
+  echo -e "${IGreen}StrVCTVRE requires hg38.phyloP100way.bw (~9.2GB). Download it from UCSC:${Color_Off}"
+  echo -e "${IGreen}  https://hgdownload.cse.ucsc.edu/goldenpath/hg38/phyloP100way/hg38.phyloP100way.bw${Color_Off}"
+  echo -e "${On_Red}\tMount it into the container (e.g. docker: -v /host/data; apptainer: --bind /host/data)${Color_Off}"
+  echo -e "${On_Red}\tor pass its path to StrVCTVRE.py with -p.${Color_Off}"
+fi
+
+# Run whatever command was passed to the container (e.g. python StrVCTVRE.py ...),
+# or drop into an interactive shell if none was given.
+if [ "$#" -gt 0 ]; then
+  source /usr/local/bin/_activate_current_env.sh
+  exec "$@"
+else
+  exec /bin/bash
+fi