Add/update software (#322)

espenhgn · web-flow · commit 5f7d420197cc · 2026-02-23T13:53:21.000+01:00
* Add/update software Fixes #315 * added SMR * offline build info * remove deprecated R raveio package
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -37,6 +37,13 @@ Removed features goes here
 
 Miscellaneous goes here
 
+## 1.15.0 - 2026-02-23
+
+### Added
+
+* Added `OPERA` and `SMR` tools to `gwas.sif` container, with corresponding tests.
+* Added Python and R packages to `python3.sif` and `r.sif` containers, with corresponding tests.
+
 ## 1.14.0 - 2025-12-05
 
 ### Updated
diff --git a/docker/README.md b/docker/README.md
@@ -17,6 +17,23 @@ The build process is defined in the corresponding [.github/workflows/docker_buil
 In general, a created PR will trigger the build process and run the unit tests automatically.
 Resulting container files will be uploaded to the [GitHub packages](https://github.com/orgs/comorment/packages?repo_name=containers) when new Git tags are pushed on the main branch.
 
+### Offline builds
+
+It should be possible to build the Docker containers offline, using the provided Dockerfiles.
+Suggested steps for building the containers offline are as follows:
+
+```{bash}
+docker build --platform=linux/amd64 -t ghcr.io/comorment/<image> -f dockerfiles/<image>/Dockerfile .
+```
+
+To successfully build the `r` container, a GITHUB_PAT file with a valid GitHub Personal Access Token (PAT) is required, which can be created in the GitHub account settings and should have at least `read:packages` scope.
+This is needed to access a number of R package GitHub repositories.
+
+```{bash}
+docker build --secret id=github_pat,src=~/secrets/GITHUB_PAT --platform=linux/amd64 -t ghcr.io/comorment/r -f dockerfiles/r/Dockerfile .
+```
+
+
 ## Note about NREC machine (old)
 
 We use NREC machine to develop and build containers.
diff --git a/docker/dockerfiles/gwas/Dockerfile b/docker/dockerfiles/gwas/Dockerfile
@@ -202,4 +202,18 @@ COPY /scripts/install_liftover.sh /tmp/liftover/
 RUN bash /tmp/liftover/install_liftover.sh && \
     rm -rf /tmp/liftover
 
+# OPERA
+WORKDIR /tmp/opera
+RUN wget --progress=dot:giga https://github.com/wuyangf7/OPERA/blob/main/opera_Linux.zip?raw=true -O opera_Linux.zip && \
+    unzip opera_Linux.zip && \
+    rm opera_Linux.zip && \
+    mv opera_Linux/opera_Linux /usr/bin/opera
+
+# SMR
+WORKDIR /tmp/smr
+RUN wget --progress=dot:giga https://yanglab.westlake.edu.cn/software/smr/download/smr-1.4.0-linux-x86_64.zip && \
+    unzip smr-1.4.0-linux-x86_64.zip && \
+    rm smr-1.4.0-linux-x86_64.zip && \
+    mv smr-1.4.0-linux-x86_64/smr /usr/bin/smr
+
 WORKDIR /tools
diff --git a/docker/dockerfiles/python3/Dockerfile b/docker/dockerfiles/python3/Dockerfile
@@ -22,14 +22,21 @@ RUN conda install mamba -n base -c conda-forge && \
 # keep the list of packages sorted alphabetically
 # https://www.online-utility.org/text/sort.jsp
 RUN mamba install python=3.12.12 \
+    BeautifulSoup4=4.14.3 \
+    bottleneck=1.6.0 \
     configparser=7.2.0 \
     dask=2025.11.0 \
     fastparquet=2024.11.0 \
+    fsspec=2026.2.0 \
+    gcsfs=2026.2.0 \
     graphviz=14.0.5 \
     "h5py=3.15.1=nompi*" \
+    html5lib=1.1 \
+    Jinja2=3.1.6 \
     jupyterlab=4.5.0 \
     lightgbm=4.6.0 \
     lifelines=0.30.0 \
+    lxml=6.0.2 \
     imbalanced-learn=0.14.0 \
     intervaltree=3.1.0 \
     matplotlib-venn=1.1.2 \
@@ -40,27 +47,42 @@ RUN mamba install python=3.12.12 \
     networkx=3.6 \
     numdifftools=0.9.41 \
     numba=0.62.1 \
+    numexpr=2.14.1 \
     numpy=2.3.5 \
+    odfpy=1.4.1 \
     openpyxl=3.1.5 \
     pandas=2.3.3 \
+    pandas-gbq=0.28.0 \
     pandas-plink=2.3.2 \
     paramiko=4.0.0 \
     psutil=7.1.3 \
+    psycopg2=2.9.11 \
     pyarrow=22.0.0 \
     pydot=4.0.1 \
+    pymysql=1.1.2 \
+    pyqt=5.15.11 \
+    pyreadr=0.5.4 \
     pyreadstat=1.3.2 \
     pytables=3.10.2 \
     python-graphviz=0.21 \
+    python-blosc2=4.0.0 \
+    pyxlsb=1.0.10 \
     pyyaml=6.0.3 \
+    qtpy=2.4.3 \
+    s3fs=2026.2.0 \
     scikit-learn=1.7.2 \
     scikit-survival=0.25.0 \
     scipy=1.16.3 \
     seaborn=0.13.2 \
     semantic_version=2.10.0 \
     shap=0.48.0 \
     statsmodels=0.14.5 \
+    SQLAlchemy=2.0.46 \
+    tabulate=0.9.0 \
+    xarray=2026.2.0 \
     xgboost=3.1.2 \
     xlrd=2.0.2 \
+    xlsxwriter=3.2.9 \
     xmltodict=1.0.2 \
     --yes && \
     mamba clean -a -y
diff --git a/docker/scripts/R/bioconductor.R b/docker/scripts/R/bioconductor.R
@@ -23,6 +23,7 @@ packages <- c(
     'GWASTools',
     'limma',
     'org.Hs.eg.db',
+    'rhdf5',
     'rtracklayer',
     'snpStats',
     'survcomp',
diff --git a/docker/scripts/R/cran.R b/docker/scripts/R/cran.R
@@ -72,6 +72,8 @@ packages <- list(
     'lightgbm',
     'lmerTest',
     'magrittr',
+    'matlab',
+    'matlab2r',
     'MatrixEQTL',
     'matrixStats',
     'MendelianRandomization',
@@ -97,6 +99,8 @@ packages <- list(
     'remotes',
     'RiskScorescvd',
     'rjags',
+    'rmatio',
+    'R.matlab',
     'ROCR',
     'rmarkdown',
     'rstan',
@@ -120,6 +124,7 @@ packages <- list(
     'vctrs',
     'xgboost',
     'zlib')
+
  
 # install package from CRAN and quit with error if installation fails
 # for (package in names(packages)) {
diff --git a/docker/scripts/R/github.R b/docker/scripts/R/github.R
@@ -26,6 +26,7 @@ packages <- list(
     "qingyuanzhao/mr.raps" = "27b96f31e26ae97a395422bd757f514a7f96cc85",
     "MRCIEU/TwoSampleMR" = "578c68fa754c57d764553812bf85d69ecf43b011",
     "MRCIEU/gwasglue" = "c2d5660eed389e1a9b3e04406b88731d642243f1",
+    # misc
     "noahlorinczcomi/MRBEE" = "6295549a1f5a158c6701eb793646d60c8aef11ca",
     "norment/normentR" = "dfa1fbae9587db6c3613b0405df4f9cfa98ee0e1",
     "psychgen/phenotools" = "62dd11e111d8d952837c9f207557e9b297ba56bc",
@@ -38,7 +39,9 @@ packages <- list(
     "JBPG/Gsens" = "6cac02ba1ccaf38870e2526076f0306c0cf0dc0a",
     "MRCIEU/genetics.binaRies" = "2fcd3ee3088b729c7eb34cf2aac9dc2e04fe4412",
     "mrcieu/gwasglue2" = "c93b3a1fca7d2eae5d40bd62117091b1ad57f0fa",
-    "gqi/MRMix" = "56afdb2bc96760842405396f5d3f02e60e305039")
+    "gqi/MRMix" = "56afdb2bc96760842405396f5d3f02e60e305039",
+    "JianYang-Lab/gsmr2" = "e7953cc136ad1cfae14674facde21e90caddea5f",
+    "n-mounier/MRlap" = "660f026864f8bfbbad5a8206bdff7d58f5d5d05b")
 
 # install package from GitHub and quit with error if installation fails
 for (package in names(packages)) {
diff --git a/docker/scripts/apt_get_essential.sh b/docker/scripts/apt_get_essential.sh
@@ -20,7 +20,7 @@ apt-get update && apt-get install -y --no-install-recommends \
    less=590-2ubuntu2.1 \
    libatlas-base-dev=3.10.3-13ubuntu1 \
    libcurl4-openssl-dev=8.5.0-2ubuntu10.6 \
-   libgomp1=14.2.0-4ubuntu2~24.04 \
+   libgomp1=14.2.0-4ubuntu2~24.04.1 \
    libgsl-dev=2.7.1+dfsg-6ubuntu2 \
    libnss3=2:3.98-1build1 \
    libpcre2-dev=10.42-4ubuntu2.1 \
diff --git a/docker/scripts/install_eagle.sh b/docker/scripts/install_eagle.sh
@@ -2,7 +2,7 @@
 set -euo pipefail
 
 apt-get update && apt-get install --no-install-recommends \
-    libopenblas-dev=0.3.26+ds-1 \
+    libopenblas-dev=0.3.26+ds-1ubuntu0.1 \
     -y && \
     apt-get clean && \
     rm -rf /var/lib/apt/lists/*
diff --git a/docker/scripts/install_htslib.sh b/docker/scripts/install_htslib.sh
@@ -5,7 +5,7 @@ set -euo pipefail
 apt-get update && apt-get install --no-install-recommends \
     libbz2-dev=1.0.8-5.1build0.1 \
     liblzma-dev=5.6.1+really5.4.5-1ubuntu0.2 \
-    libssl-dev=3.0.13-0ubuntu3.6 \
+    libssl-dev=3.0.13-0ubuntu3.7 \
     -y 
  
 apt-get clean && \
diff --git a/docker/scripts/install_switchError.sh b/docker/scripts/install_switchError.sh
@@ -3,7 +3,7 @@ set -euo pipefail
 
 apt-get update && apt-get install --no-install-recommends \
     r-mathlib=4.3.3-2build2 \
-    libboost-iostreams1.83-dev=1.83.0-2.1ubuntu3.1 \
+    libboost-iostreams1.83-dev=1.83.0-2.1ubuntu3.2 \
     -y && \
     apt-get clean && \
     rm -rf /var/lib/apt/lists/*
diff --git a/docs/containers/gwas.md b/docs/containers/gwas.md
@@ -48,6 +48,7 @@ List of software included in the container:
   | liftOver[^liftover]             | latest                                    | [permissive](https://genome-store.ucsc.edu)
   | metal[^metal]                   | 2020-05-05                                | -
   | minimac4[^minimac4]             | v4.1.6                                    | [GPLv3](https://www.gnu.org/licenses/gpl-3.0.html)
+  | OPERA[^opera]                   | v1.18                                     | -
   | plink[^plink19]                 | v1.9.0-b.7.11 64-bit (19 Aug 2025)            | [GPLv3](https://www.gnu.org/licenses/gpl-3.0.html)
   | plink2[^plink2]                 | v2.0.0-a.6.23LM 64-bit Intel (24 Aug 2025)    | [GPLv3](https://www.gnu.org/licenses/gpl-3.0.html)
   | plink2_avx2[^plink2]            | v2.0.0-a.6.23LM AVX2 Intel (24 Aug 2025)      | [GPLv3](https://www.gnu.org/licenses/gpl-3.0.html)
@@ -57,6 +58,7 @@ List of software included in the container:
   | samtools[^bcftools]             | v1.19.2                                   | [MIT/ExpatD](https://github.com/samtools/samtools/blob/develop/LICENSE)
   | shapeit4.2[^shapeit4]           | v4.2.2                                    | [MIT](https://opensource.org/licenses/MIT)
   | simu_linux[^simu]               | v0.9.4                                    | [GPLv3](https://www.gnu.org/licenses/gpl-3.0.html)
+  | SMR[^smr]                       | v1.3.2                                    | [GPLv2](https://www.gnu.org/licenses/gpl-2.0.html)
   | snptest[^snptest]               | v2.5.6                                    | [permissive](https://www.chg.ox.ac.uk/~gav/snptest/#download)
   | switchError[^switcherror]       | 6e688b1                                   | [MIT](https://opensource.org/licenses/MIT)
   | vcftools[^vcftools]             | 0.1.17 (git SHA: d511f469e)               | [GPLv3](https://www.gnu.org/licenses/gpl-3.0.html)
@@ -101,6 +103,8 @@ List of software included in the container:
 
 [^minimac4]: [https://genome.sph.umich.edu/wiki/Minimac4](https://genome.sph.umich.edu/wiki/Minimac4)
 
+[^opera]: Wu Y., Qi T., Wray N.R., Visscher P.M., Zeng J. & Yang J. (2023) Joint analysis of GWAS and multi-omics QTL summary statistics reveals a large fraction of GWAS signals shared with molecular phenotypes. Cell Genomics.
+
 [^plink19]: Christopher C Chang, Carson C Chow, Laurent CAM Tellier, Shashaank Vattikuti, Shaun M Purcell, James J Lee, Second-generation PLINK: rising to the challenge of larger and richer datasets, GigaScience, Volume 4, Issue 1, December 2015, s13742–015–0047–8, <https://doi.org/10.1186/s13742-015-0047-8>
 
 [^plink2]: [https://www.cog-genomics.org/plink/2.0/](https://www.cog-genomics.org/plink/2.0/)
@@ -115,6 +119,8 @@ List of software included in the container:
 
 [^simu]: [https://github.com/precimed/simu](https://github.com/precimed/simu)
 
+[^smr]: Zhu Z, Zhang F, Hu H, Bakshi A, Robinson MR, Powell JE, Montgomery GW, Goddard ME, Wray NR, Visscher PM & Yang J (2016) Integration of summary data from GWAS and eQTL studies predicts complex trait gene targets. Nature Genetics, 48:481-487.
+
 [^snptest]: [https://www.chg.ox.ac.uk/~gav/snptest/#download](https://www.chg.ox.ac.uk/~gav/snptest/#download)
 
 [^switcherror]: O'Connell J, Gurdasani D, Delaneau O, Pirastu N, Ulivi S, et al. (2014) A General Approach for Haplotype Phasing across the Full Spectrum of Relatedness. PLOS Genetics 10(4): e1004234. <https://doi.org/10.1371/journal.pgen.1004234>
diff --git a/tests/extras/r.R b/tests/extras/r.R
@@ -106,6 +106,7 @@ libraries_to_check <- c(
     "gplots",
     "gridExtra",
     "Gsens",
+    "gsmr2",
     "gwasglue",
     "gwasglue2",
     "gwasurvivr",
@@ -126,12 +127,15 @@ libraries_to_check <- c(
     "lubridate",
     "kinship2",
     "MASS",
+    "matlab",
+    "matlab2r",
     "MatrixEQTL",
     "matrixStats",
     "MendelianRandomization",
     "mgcv",
     "miniCRAN",
     "MRBEE",
+    "MRlap",
     "MVMR",
     "modelr",
     "moments",
@@ -150,6 +154,9 @@ libraries_to_check <- c(
     "pracma",
     "PredictABEL",
     "psych",
+    "rhdf5",
+    "R.matlab",
+    "rmatio",
     "RColorBrewer",
     "Rcpp",
     "readxl",
@@ -177,6 +184,7 @@ libraries_to_check <- c(
     "tree",
     "viridis")
 
+
 # Define a test
 test_that("Required libraries are installed", {
   missing_libraries <- setdiff(libraries_to_check, installed.packages()[, "Package"])
diff --git a/tests/test_gwas.py b/tests/test_gwas.py
@@ -37,7 +37,6 @@
             f'--mount type=bind,source={cwd},target={cwd} ' +
             '{custom_mount}' +
             '-w /home/ ' +
-            # '--platform linux/amd64 ' +
             'ghcr.io/comorment/gwas')
     except FileNotFoundError as err:
         # neither singularity nor docker found, fall back to plain python
@@ -339,3 +338,16 @@ def test_gwas_duohmm():
     call = f'{PREFIX} duohmm'
     out = subprocess.run(call.split(' '), check=False)
     assert out.returncode == 0
+
+
+def test_gwas_opera():
+    """test opera"""
+    call = f'{PREFIX} opera'
+    out = subprocess.run(call.split(' '), shell=True, check=False, capture_output=True)
+    assert out.returncode == 0
+
+def test_gwas_smr():
+    """test smr"""
+    call = f'{PREFIX} smr'
+    out = subprocess.run(call.split(' '), shell=True, check=False, capture_output=True)
+    assert out.returncode == 0
diff --git a/tests/test_python3.py b/tests/test_python3.py
@@ -106,42 +106,63 @@ def test_python3_miniwdl():
 
 def test_python3_packages():
     packages = [
+        'blosc2',
+        'bottleneck',
+        'bs4',  # BeautifulSoup4
         'configparser',
         'crc32c',
         'dask',
-        'fastparquet',
         'dxpy',
+        'fastparquet',
+        'fsspec',
+        'gcsfs',
         'graphviz',
         'h5py',
+        'html5lib',
         'imblearn',
         'intervaltree',
         'ldpred',
         'lifelines',
         'lightgbm',
+        'lxml',
         'matplotlib',
         'matplotlib_venn',
         'numba',
         'numdifftools',
         'numpy',
+        'numexpr',
+        'odf',
         'openpyxl',
         'pandas',
+        'pandas_gbq',
         'pandas_plink',
         'paramiko',
         'plinkio',
+        'psycopg2',
         'pyarrow',
         'pydot',
         'pyliftover',
+        'pymysql',
+        'PyQt5',
+        'pyreadr',
         'pyreadstat',
+        'pyxlsb',
+        'qtpy',
         'redcap',  # pycap
         'scipy',
+        's3fs',
         'seaborn',
         'semantic_version',
         'shap',
         'sklearn',
         'sksurv',
+        'sqlalchemy',
         'statsmodels',
         'tables',
+        'tabulate',
+        'xarray',
         'xlrd',
+        'xlsxwriter',
         'xmltodict',
         'xgboost',
         'yaml',  # pyyaml
diff --git a/version/version.py b/version/version.py
@@ -1,5 +1,5 @@
 _MAJOR = "1"
-_MINOR = "14"
+_MINOR = "15"
 # On main and in a nightly release the patch should be one ahead of the last
 # released build.
 _PATCH = "0"
