Merge branch 'pdb_ihm_rename'

README.md

@@ -1,7 +1,7 @@
 # Validation pipeline for integrative and hybrid models
 
 ## Project objective 
-- `1.` Develop a validation pipeline (including data validation, model validation, fit of input to model, fit of data not used for modeling and uncertainty of the model) for assessing IHM structures deposited to [PDB-Dev](https://pdb-dev.wwpdb.org/index.html)
+- `1.` Develop a validation pipeline (including data validation, model validation, fit of input to model, fit of data not used for modeling and uncertainty of the model) for assessing IHM structures deposited to [PDB-IHM](https://pdb-ihm.org/index.html)
 
 ## List of files and directories:
 - `docs` documentation for all classes and functions (sphinx)

docker/Dockerfile

@@ -1,23 +1,43 @@
-FROM centos:7
-MAINTAINER Ben Webb <[email protected]>
+FROM ubuntu:22.04
+MAINTAINER Arthur Zalevsky <[email protected]>
 
-# Install ATSAS and its dependencies
-COPY ATSAS-3.0.5-2.el7.x86_64.rpm /
-RUN yum -y localinstall /ATSAS-3.0.5-2.el7.x86_64.rpm && yum -y install libgfortran5 git && rm -f /ATSAS*.rpm
+ARG ATSAS
+ARG TZ=America/Los_Angeles
+ENV TZ=${TZ}
+ENV DEBIAN_FRONTEND=noninteractive
 
-# Set up basic conda environment using conda-forge packages
-RUN curl -L https://github.com/conda-forge/miniforge/releases/download/4.11.0-0/Miniforge3-4.11.0-0-Linux-x86_64.sh -o miniforge.sh && echo "4c24b38969ac413efa3a984290296f72578340d06004f2c7ba5efcbf828ec86b  miniforge.sh" sha256sum -c --status && bash ./miniforge.sh -b -p /root/miniforge && rm -f miniforge.sh && /root/miniforge/bin/conda init -q bash
+# Install common dependencies
+RUN apt-get update -y && apt-get install -y curl git subversion build-essential
 
-# Install Python packages needed by IHMValidation and install extra RPMs
-# needed to run firefox
-RUN source /root/.bashrc && eval "$(conda shell.bash hook)" && conda install -y pandas flask ihm scikit-learn selenium bokeh requests firefox geckodriver && pip install python-decouple pdfkit && yum -y install gtk3 alsa-lib
+# Set up basic conda environment using conda-forge packages
+RUN curl -L https://github.com/conda-forge/miniforge/releases/download/23.3.1-1/Mambaforge-23.3.1-1-Linux-x86_64.sh -o miniforge.sh && echo "dbe92c011a1315b9626e2f93a165892f4b89177145bc350b3859a483a3642a24  miniforge.sh" sha256sum -c --status && bash ./miniforge.sh -b -p /opt/conda && rm -f miniforge.sh && /opt/conda/bin/conda init -q bash
 
-# Get wkhtmltopdf directly from web site (there is a conda package, but it
-# does not handle https: links)
-RUN curl -LO https://github.com/wkhtmltopdf/packaging/releases/download/0.12.6-1/wkhtmltox-0.12.6-1.centos7.x86_64.rpm && yum -y localinstall wkhtmltox-0.12.6-1.centos7.x86_64.rpm && rm -f wkhtmltox-0.12.6-1.centos7.x86_64.rpm
+# Ensure that we always use a shell with conda
+ENV PATH /opt/conda/bin:${PATH}
+RUN echo "conda activate" >> ~/.bashrc
+SHELL ["/bin/bash", "--login", "-c"] 
 
 # Build molprobity from source code and put the binaries we need in /usr/bin/
-RUN source /root/.bashrc && eval "$(conda shell.bash hook)" && yum -y install mesa-libGL-devel subversion ca-certificates git gcc-c++ && mkdir molprobity && cd molprobity && curl -O https://raw.githubusercontent.com/rlabduke/MolProbity/master/install_via_bootstrap.sh && bash install_via_bootstrap.sh 12 && ln -sf /molprobity/build/bin/molprobity.clashscore /usr/bin/ && ln -sf /molprobity/build/bin/molprobity.ramalyze /usr/bin/ && ln -sf /molprobity/build/bin/molprobity.rotalyze /usr/bin/ && ln -sf /molprobity/build/bin/molprobity.molprobity /usr/bin/
+ADD "https://api.github.com/repos/rlabduke/MolProbity/commits?per_page=1" latest_commit
+RUN mkdir /opt/molprobity && cd /opt/molprobity && curl -O https://raw.githubusercontent.com/rlabduke/MolProbity/master/install_via_bootstrap.sh && bash ./install_via_bootstrap.sh 4
+ENV PATH ${PATH}:/opt/molprobity/build/bin
+
+# Install SAS tools: ATSAS and SASCIFTOOLS
+COPY ${ATSAS} /.
+RUN apt-get update -y && apt-get install -y /${ATSAS} && rm -f /${ATSAS}
+RUN pip install git+https://git.embl.de/grp-svergun/sasciftools3.git@master#egg=sasciftools
+# Install additional certificate for SASBDB
+RUN curl https://cacerts.digicert.com/ThawteEVRSACAG2.crt.pem -o ThawteEVRSACAG2.crt --output-dir /usr/local/share/ca-certificates/ && update-ca-certificates
+ENV REQUESTS_CA_BUNDLE=/etc/ssl/certs/ca-certificates.crt
+
+# Install Python packages and other dependencies needed by IHMValidation
+RUN apt-get update -y && apt-get install -y xvfb libasound2 libdbus-glib-1-2 libgtk-3-0 && mamba install -y pip requests numpy pandas scikit-learn matplotlib "bokeh<3.0" selenium pyvirtualdisplay firefox geckodriver jinja2 && pip install -U iqplot pdfkit ihm && curl -L -O https://github.com/wkhtmltopdf/packaging/releases/download/0.12.6.1-2/wkhtmltox_0.12.6.1-2.jammy_amd64.deb && apt install -y ./wkhtmltox_0.12.6.1-2.jammy_amd64.deb && rm -rf wkhtmltox_0.12.6.1-2.jammy_amd64.deb
+# Hardcode selenium path
+ENV SE_MANAGER_PATH=/opt/conda/bin/selenium-manager
+
+# Get IHMValidation
+ADD "https://api.github.com/repos/salilab/IHMValidation/commits?per_page=1" latest_commit
+RUN git clone --depth 1 https://github.com/salilab/IHMValidation.git /opt/IHMValidation
 
-# Get IHMValidation and configure
-RUN git clone https://github.com/salilab/IHMValidation.git && source /root/.bashrc && eval "$(conda shell.bash hook)" && yum -y install which && echo "ATSAS=\"`which datcmp`\"" > IHMValidation/.env && echo "Molprobity_ramalyze=\"`which molprobity.ramalyze`\"" >> IHMValidation/.env && echo "Molprobity_molprobity=\"`which molprobity.molprobity`\"" >> IHMValidation/.env && echo "Molprobity_clashscore=\"`which molprobity.clashscore`\"" >> IHMValidation/.env && echo "Molprobity_rotalyze=\"`which molprobity.rotalyze`\"" >> IHMValidation/.env && echo "wkhtmltopdf=\"`which wkhtmltopdf`\"" >> IHMValidation/.env
+# Create input and output dirs
+RUN mkdir -p /ihmv/input /ihmv/output /ihmv/cache /ihmv/databases

docker/README.md

@@ -0,0 +1,16 @@
+# Building runtime image
+
+We switched from Docket to [Singularity](https://docs.sylabs.io/guides/4.1/user-guide/introduction.html) containers to run the validation pipeline. 
+
+To build the image:
+1. Navigate to the directory holding `Singularity.def` file:
+`cd IHMValidation/docker`
+2. Download [ATSAS](https://www.embl-hamburg.de/biosaxs/download.html) `.deb` package for Ubuntu 22.04 and put into the same directory. The current pipeline was tested with `ATSAS-3.2.1` but newer versions should work too.
+3. Select an appropriate [timezone](https://en.wikipedia.org/wiki/List_of_tz_database_time_zones). For example `America/Los_Angeles`
+4. Build the image:
+`singularity build --build-arg TZ=America/Los_Angeles --build-arg ATSAS=./ATSAS-3.2.1-1_amd64.deb ihmv_$(date +%Y%m%d).sif Singularity.def`
+which will generate a `.sif` image using current date as a nametag (i.e. `ihmv_20240227.sif`)
+
+**NB 1**: It takes about ~25 minutes on a modern workstation to build the image from scratch. 
+
+**NB 2**: Due to the [Molprobity](https://github.com/rlabduke/MolProbity)'s rolling release model we can't freeze Molprobity version thus we can't guarantee the `IHMValidation` package will work with a freshly-built image.