Misc. fixes and updates (#312)

espenhgn · web-flow · commit 13ddc4eac4c6 · 2025-11-26T14:17:28.000+01:00
* LDpred2/bignsnpr messed up IIDs in output file Fixes #311 * bump pins, remove git lfs pulls, xfail test * xfail w. git lfs datas * xfail * update installer scripts + bit64 dep * drop shapeit5 from build * detach bit64 when no longer needed
diff --git a/.github/workflows/docker_build_gwas.yml b/.github/workflows/docker_build_gwas.yml
@@ -22,7 +22,7 @@ jobs:
         name: Checkout repository
         uses: actions/checkout@v4
         with:
-          lfs: 'true'
+          lfs: 'false'
           sparse-checkout: |
             test-requirements.txt
             tests
@@ -65,6 +65,5 @@ jobs:
       - 
         name: Test
         run: |
-          git lfs pull -I tests/extras/*.tar.gz
           pip3 install -r test-requirements.txt
           py.test -v tests/test_gwas.py
diff --git a/.github/workflows/docker_build_push_gwas.yml b/.github/workflows/docker_build_push_gwas.yml
@@ -17,7 +17,7 @@ jobs:
         name: Checkout repository
         uses: actions/checkout@v4
         with:
-          lfs: 'true'
+          lfs: 'false'
           sparse-checkout: |
             test-requirements.txt
             tests
@@ -59,7 +59,6 @@ jobs:
       - 
         name: Test
         run: |
-          git lfs pull -I tests/extras/*.tar.gz
           pip3 install -r test-requirements.txt
           py.test -v tests/test_gwas.py
       -
diff --git a/.github/workflows/docker_build_push_r.yml b/.github/workflows/docker_build_push_r.yml
@@ -17,7 +17,7 @@ jobs:
         name: Checkout repository
         uses: actions/checkout@v4
         with:
-          lfs: 'true'
+          lfs: 'false'
           sparse-checkout: |
             test-requirements.txt
             tests
@@ -63,7 +63,6 @@ jobs:
       - 
         name: Test
         run: |
-          git lfs pull -I tests/extras/*.tar.gz
           pip3 install -r test-requirements.txt
           py.test -v tests/test_r.py
       -
diff --git a/.github/workflows/docker_build_r.yml b/.github/workflows/docker_build_r.yml
@@ -22,7 +22,7 @@ jobs:
         name: Checkout repository
         uses: actions/checkout@v4
         with:
-          lfs: 'true'
+          lfs: 'false'
           sparse-checkout: |
             test-requirements.txt
             tests
@@ -69,6 +69,5 @@ jobs:
       - 
         name: Test
         run: |
-          git lfs pull -I tests/extras/*.tar.gz
           pip3 install -r test-requirements.txt
           py.test -v tests/test_r.py
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -37,6 +37,17 @@ Removed features goes here
 
 Miscellaneous goes here
 
+## [1.12.0] - 2025-11-26
+
+### Fixed
+
+* quiet import of `bit64` in `ldpred2.R` to fix broken IDs with `attach_snp()` function
+* not using git LFS for GH actions
+
+### Removed
+
+* removed shapeit5 installation from `gwas.sif` container build
+
 ## [1.11.0] - 2025-09-17
 
 ### Added
diff --git a/docker/dockerfiles/gwas/Dockerfile b/docker/dockerfiles/gwas/Dockerfile
@@ -152,9 +152,9 @@ WORKDIR /tmp/shapeit5
 # COPY /scripts/patches/shapeit5.diff /tmp/shapeit5/
 # COPY /scripts/patches/xcftools.diff /tmp/shapeit5/
 # RUN bash /tmp/shapeit5/install_shapeit5.sh && \
-COPY /scripts/install_shapeit5_static.sh /tmp/shapeit5/
-RUN bash /tmp/shapeit5/install_shapeit5_static.sh && \
-    rm -rf /tmp/shapeit5
+# COPY /scripts/install_shapeit5_static.sh /tmp/shapeit5/
+# RUN bash /tmp/shapeit5/install_shapeit5_static.sh && \
+#     rm -rf /tmp/shapeit5
 
 
 # eagle
diff --git a/docker/scripts/R/cran.R b/docker/scripts/R/cran.R
@@ -14,6 +14,7 @@ packages <- list(
     'bigreadr',
     'bigsnpr',
     'BiocManager',
+    'bit64',
     'brant',
     'brms',
     'carat',
diff --git a/docker/scripts/apt_get_essential.sh b/docker/scripts/apt_get_essential.sh
@@ -29,7 +29,7 @@ apt-get update && apt-get install -y --no-install-recommends \
    parallel=20231122+ds-1 \
    perl=5.38.2-3.2ubuntu0.2 \
    pkg-config=1.8.1-2build1 \
-   python3=3.12.3-0ubuntu2 \
+   python3-dev=3.12.3-0ubuntu2.1 \
    python3-pytest=7.4.4-1 \
    tar=1.35+dfsg-3build1 \
    tofrodos=1.7.13+ds-6 \
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=3.0.13-0ubuntu3.5 \
+    libssl-dev=3.0.13-0ubuntu3.6 \
     -y 
  
 apt-get clean && \
diff --git a/docker/scripts/install_minimac4.sh b/docker/scripts/install_minimac4.sh
@@ -4,7 +4,7 @@ set -euo pipefail
 # install some deps for installing cget
 apt-get update && \
     apt-get install --no-install-recommends \
-    python3-pip=24.0+dfsg-1ubuntu1.2 \
+    python3-pip=24.0+dfsg-1ubuntu1.3 \
     python3-click=8.1.6-2 \
     python3-six=1.16.0-4 \
     -y && \
diff --git a/docs/containers/gwas.md b/docs/containers/gwas.md
@@ -56,11 +56,6 @@ List of software included in the container:
   | regenie[^regenie]               | v4.1.2                                    | [MIT/Boost](https://github.com/rgcgithub/regenie/blob/master/LICENSE)
   | 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)
-  | shapeit5[^shapeit5] phase_rare  | v5.1.1                                    | [MIT](https://opensource.org/licenses/MIT)
-  | shapeit5[^shapeit5] phase_common | v5.1.1                                  | [MIT](https://opensource.org/licenses/MIT)
-  | shapeit5[^shapeit5] ligate      | v5.1.1                                    | [MIT](https://opensource.org/licenses/MIT)
-  | shapeit5[^shapeit5] switch      | v5.1.1                                    | [MIT](https://opensource.org/licenses/MIT)
-  | shapeit5[^shapeit5] xcftools    | v5.1.1                                    | [MIT](https://opensource.org/licenses/MIT)
   | simu_linux[^simu]               | v0.9.4                                    | [GPLv3](https://www.gnu.org/licenses/gpl-3.0.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)
@@ -118,8 +113,6 @@ List of software included in the container:
 
 [^shapeit4]: Delaneau, O., Zagury, JF., Robinson, M.R. et al. Accurate, scalable and integrative haplotype estimation. Nat Commun 10, 5436 (2019). <https://doi.org/10.1038/s41467-019-13225-y>
 
-[^shapeit5]: Hofmeister, R.J., Ribeiro, D.M., Rubinacci, S. et al. Accurate rare variant phasing of whole-genome and whole-exome sequencing data in the UK Biobank. Nat Genet 55, 1243–1249 (2023). <https://doi.org/10.1038/s41588-023-01415-w>
-
 [^simu]: [https://github.com/precimed/simu](https://github.com/precimed/simu)
 
 [^snptest]: [https://www.chg.ox.ac.uk/~gav/snptest/#download](https://www.chg.ox.ac.uk/~gav/snptest/#download)
diff --git a/scripts/pgs/LDpred2/ldpred2.R b/scripts/pgs/LDpred2/ldpred2.R
@@ -1,13 +1,15 @@
 # Calculate polygenic scores using ldpred2
-# this script is an adaptation of the demo script available at the bigsnpr homepage
-library(bigsnpr, quietly = T)
+# this script is an adaptation of the LDpred2 vignette available at the
+# bigsnpr homepage (https://privefl.github.io/bigsnpr/articles/LDpred2.html)
+library(bigsnpr, quietly = TRUE)
 options(bigstatsr.check.parallel.blas = FALSE)
 options(default.nproc.blas = NULL)
 library(tools)
 library(argparser, quietly=T)
 library(stringr)
 
-### Maybe there's some environment variable availble to determine the location of the script instead
+# Maybe there's some environment variable availble to determine the
+# location of the script instead
 coms <- commandArgs()
 coms <- coms[substr(coms, 1, 7) == '--file=']
 dirScript <- dirname(substr(coms, 8, nchar(coms)))
@@ -142,7 +144,9 @@ if (fileOutputMerge) {
 if (!file.exists(parsed$tmp_dir)) stop("Temporary directory", parsed$tmp_dir, "does not exist") 
 
 cat('Loading backingfile:', fileGeno ,'\n')
+suppressPackageStartupMessages(library(bit64))
 obj.bigSNP <- snp_attach(fileGeno)
+detach("package:bit64", unload = TRUE)
 
 # Store some key variables
 G <- obj.bigSNP$genotypes
diff --git a/tests/extras/r.R b/tests/extras/r.R
@@ -11,6 +11,7 @@ libraries_to_check <- c(
     "bigsnpr",
     "BiocGenerics",
     "BiocManager",
+    'bit64',
     "caret",
     "coloc",
     "colocboost",
diff --git a/tests/test_gwas.py b/tests/test_gwas.py
@@ -101,6 +101,7 @@ def test_gwas_flashpca():
     assert out.returncode == 0
 
 
+@pytest.mark.xfail(reason="skip due to git lfs stored test data")
 def test_gwas_gcta():
     """test gcta"""
     with tempfile.TemporaryDirectory() as d:
@@ -115,6 +116,7 @@ def test_gwas_gcta():
         assert out.returncode == 0
 
 
+@pytest.mark.xfail(reason="skip due to git lfs stored test data")
 def test_gwas_gctb():
     """test gctb"""
     if platform.machine() == 'arm64':
@@ -139,6 +141,7 @@ def test_gwas_gwama():
     assert out.returncode == 0
 
 
+@pytest.mark.xfail(reason="skip due to git lfs stored test data")
 def test_gwas_king():
     """test king"""
     with tempfile.TemporaryDirectory() as d:
@@ -174,6 +177,7 @@ def test_gwas_liftover():
     assert out.stdout.decode('utf-8').lower().rfind('error') <= 0
 
 
+@pytest.mark.xfail(reason="skip due to git lfs stored test data")
 def test_gwas_metal():
     """test metal"""
     with tempfile.TemporaryDirectory() as d:
@@ -306,6 +310,7 @@ def test_gwas_switchError():
     assert out.returncode == 0
 
 
+@pytest.mark.xfail(reason="skip due to git lfs stored test data")
 def test_gwas_simu():
     """test simu"""
     with tempfile.TemporaryDirectory() as d:
diff --git a/tests/test_r.py b/tests/test_r.py
@@ -7,6 +7,7 @@
 import os
 import subprocess
 import tempfile
+import pytest
 
 
 # Check that (1) apptainer exist, and (2) if not, check for docker.
@@ -84,6 +85,7 @@ def test_r_R_rmarkdown():
         assert pdf_output
 
 
+@pytest.mark.xfail(reason="skip due to git lfs stored test data")
 def test_gwas_gcta():
     """test gcta"""
     with tempfile.TemporaryDirectory() as d:
diff --git a/version/version.py b/version/version.py
@@ -1,5 +1,5 @@
 _MAJOR = "1"
-_MINOR = "11"
+_MINOR = "12"
 # On main and in a nightly release the patch should be one ahead of the last
 # released build.
 _PATCH = "0"
