Merge pull request #15 from PMCC-BioinformaticsCore/tests/wgsgermline…

…gatk Tests/wgsgermlinegatk
PMCC-BioinformaticsCore · Mar 31, 2021 · 5d5c01e · 5d5c01e
2 parents c1aeec7 + 326fef8
commit 5d5c01e
Show file tree

Hide file tree

Showing 8 changed files with 91 additions and 47 deletions.
diff --git a/.github/workflows/publish.yml b/.github/workflows/publish.yml
@@ -0,0 +1,28 @@
+name: Upload Python Package
+
+on:
+  push:
+    tags:
+      - v*
+
+jobs:
+  deploy:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v2
+    - name: Set up Python
+      uses: actions/setup-python@v2
+      with:
+        python-version: '3.8'
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install .
+        pip install -e .[ci]
+    - name: Build and publish
+      env:
+        TWINE_USERNAME: ${{ secrets.PYPI_USERNAME }}
+        TWINE_PASSWORD: ${{ secrets.PYPI_PASSWORD }}
+      run: |
+        python setup.py sdist bdist_wheel
+        twine upload dist/*
diff --git a/.github/workflows/unit_tests.yml b/.github/workflows/unit_tests.yml
@@ -0,0 +1,22 @@
+name: Unit Tests
+
+on: [push]
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: [3.6, 3.7, 3.8]
+
+    steps:
+    - uses: actions/checkout@v2
+    - name: Set up Python ${{ matrix.python-version }}
+      uses: actions/setup-python@v2
+      with:
+        python-version: ${{ matrix.python-version }}
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install .
+    - uses: codecov/codecov-action@v1
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -3,10 +3,9 @@ repos:
     rev: stable
     hooks:
     - id: black
-      language_version: python3.7
-#- repo: local
-#  hooks:
-#    - id: nosetests
-#      name: nosetests
-#      entry: sh -ec 'nosetests -w janis'
-#      language: system
+- repo: local
+  hooks:
+    - id: nosetests
+      name: nosetests
+      entry: sh -ec 'nosetests -w janis_pipelines'
+      language: system
diff --git a/.travis.yml b/.travis.yml
diff --git a/janis_pipelines/__meta__.py b/janis_pipelines/__meta__.py
@@ -1,2 +1,2 @@
-__version__ = "v0.11.0"
+__version__ = "v0.11.1"
 description = "Bioinformatics tools for Janis; the Pipeline creation helper"
diff --git a/janis_pipelines/wgs_germline/wgsgermline.py b/janis_pipelines/wgs_germline/wgsgermline.py
@@ -110,35 +110,37 @@ def bind_metadata(self):
 
     def tests(self) -> Optional[List[TTestCase]]:
         bioinf_base = "https://swift.rc.nectar.org.au/v1/AUTH_4df6e734a509497692be237549bbe9af/janis-test-data/bioinformatics"
+        chr17 = f"{bioinf_base}/petermac_testdata"
         hg38 = f"{bioinf_base}/hg38"
 
         return [
             TTestCase(
                 name="brca1",
                 input={
                     "sample_name": "NA12878",
-                    "reference": f"{hg38}/reference/Homo_sapiens_assembly38.fasta",
+                    "reference": f"{chr17}/Homo_sapiens_assembly38.chr17.fasta",
                     "fastqs": [
                         [
-                            f"{bioinf_base}/BRCA1_R1.fastq.gz",
-                            f"{bioinf_base}/BRCA1_R2.fastq.gz",
+                            f"{chr17}/NA12878-BRCA1_R1.fastq.gz",
+                            f"{chr17}/NA12878-BRCA1_R2.fastq.gz",
                         ]
                     ],
-                    "gatk_intervals": [f"{hg38}/NA12878/BRCA1.bed"],
+                    "gatk_intervals": [f"{chr17}/BRCA1.hg38.bed"],
                     "vardict_intervals": [f"{hg38}/NA12878/BRCA1.intersect.bed"],
-                    "strelka_intervals": f"{hg38}/hg38.bed.gz",
-                    "known_indels": f"{hg38}/known_indels/Homo_sapiens_assembly38.known_indels.vcf.gz",
-                    "mills_indels": f"{hg38}/mills_indels/Mills_and_1000G_gold_standard.indels.hg38.vcf.gz",
-                    "snps_1000gp": f"{hg38}/snps_1000GP/1000G_phase1.snps.high_confidence.hg38.vcf.gz",
-                    "snps_dbsnp": f"{hg38}/snps_dbsnp/Homo_sapiens_assembly38.dbsnp138.vcf.gz",
-                    "cutadapt_adapters": "https://raw.githubusercontent.com/csf-ngs/fastqc/master/Contaminants/contaminant_list.txt",
-                    "gridss_blacklist": f"{hg38}/NA12878/BRCA1.bed",
+                    "strelka_intervals": f"{chr17}/BRCA1.hg38.bed.gz",
+                    "known_indels": f"{chr17}/Homo_sapiens_assembly38.known_indels.BRCA1.vcf.gz",
+                    "mills_indels": f"{chr17}/Mills_and_1000G_gold_standard.indels.hg38.BRCA1.vcf.gz",
+                    "snps_1000gp": f"{chr17}/1000G_phase1.snps.high_confidence.hg38.BRCA1.vcf.gz",
+                    "snps_dbsnp": f"{chr17}/Homo_sapiens_assembly38.dbsnp138.BRCA1.vcf.gz",
+                    "cutadapt_adapters": f"{chr17}/contaminant_list.txt",
+                    "gridss_blacklist": f"{chr17}/consensusBlacklist.hg38.chr17.bed",
+                    "MINIMUM_PCT": f""
                 },
                 output=[
                     TTestExpectedOutput(
                         tag="out_variants",
                         preprocessor=TTestPreprocessor.LinesDiff,
-                        file_diff_source=f"{hg38}/NA12878/brca1.germline.combined.vcf",
+                        file_diff_source=f"{chr17}/NA12878/brca1.germline.combined.vcf",
                         operator=lambda actual, expected: all(
                             a <= b for a, b in zip(actual, expected)
                         ),
@@ -148,13 +150,13 @@ def tests(self) -> Optional[List[TTestCase]]:
                         "out_variants_gridss",
                         preprocessor=TTestPreprocessor.FileContent,
                         operator=operator.eq,
-                        expected_file=f"{hg38}/NA12878/brca1.germline.gridss.vcf",
+                        expected_file=f"{chr17}/NA12878/brca1.germline.gridss.vcf",
                     ),
                     TTestExpectedOutput(
                         tag="out_bam",
                         preprocessor=TTestPreprocessor.Value,
                         operator=Bam.equal,
-                        expected_value=f"{hg38}/NA12878/NA12878.bam",
+                        expected_value=f"{chr17}/NA12878/NA12878.bam",
                     ),
                     TTestExpectedOutput(
                         tag="out_gridss_assembly",

diff --git a/janis_pipelines/wgs_germline_gatk/wgsgermlinegatk.py b/janis_pipelines/wgs_germline_gatk/wgsgermlinegatk.py
@@ -108,25 +108,26 @@ def tests(self) -> Optional[List[TTestCase]]:
 
         bioinf_base = "https://swift.rc.nectar.org.au/v1/AUTH_4df6e734a509497692be237549bbe9af/janis-test-data/bioinformatics"
         hg38 = f"{bioinf_base}/hg38"
+        chr17 = f"{bioinf_base}/petermac_testdata"
 
         return [
             TTestCase(
                 name="brca1",
                 input={
                     "sample_name": "NA12878",
-                    "reference": f"{hg38}/reference/Homo_sapiens_assembly38.fasta",
+                    "reference": f"{chr17}/Homo_sapiens_assembly38.chr17.fasta",
                     "fastqs": [
                         [
-                            f"{bioinf_base}/BRCA1_R1.fastq.gz",
-                            f"{bioinf_base}/BRCA1_R2.fastq.gz",
+                            f"{chr17}/NA12878-BRCA1_R1.fastq.gz",
+                            f"{chr17}/NA12878-BRCA1_R2.fastq.gz",
                         ]
                     ],
-                    "gatk_intervals": [f"{hg38}/NA12878/BRCA1.bed"],
-                    "known_indels": f"{hg38}/known_indels/Homo_sapiens_assembly38.known_indels.vcf.gz",
-                    "mills_indels": f"{hg38}/mills_indels/Mills_and_1000G_gold_standard.indels.hg38.vcf.gz",
-                    "snps_1000gp": f"{hg38}/snps_1000GP/1000G_phase1.snps.high_confidence.hg38.vcf.gz",
-                    "snps_dbsnp": f"{hg38}/snps_dbsnp/Homo_sapiens_assembly38.dbsnp138.vcf.gz",
-                    "cutadapt_adapters": "https://raw.githubusercontent.com/csf-ngs/fastqc/master/Contaminants/contaminant_list.txt",
+                    "gatk_intervals": [f"{chr17}/BRCA1.hg38.bed"],
+                    "known_indels": f"{chr17}/Homo_sapiens_assembly38.known_indels.BRCA1.vcf.gz",
+                    "mills_indels": f"{chr17}/Mills_and_1000G_gold_standard.indels.hg38.BRCA1.vcf.gz",
+                    "snps_1000gp": f"{chr17}/1000G_phase1.snps.high_confidence.hg38.BRCA1.vcf.gz",
+                    "snps_dbsnp": f"{chr17}/Homo_sapiens_assembly38.dbsnp138.BRCA1.vcf.gz",
+                    "cutadapt_adapters": f"{chr17}/contaminant_list.txt",
                 },
                 output=[
                     TTestExpectedOutput(

diff --git a/setup.py b/setup.py
@@ -31,6 +31,14 @@
         "janis-pipelines.bioinformatics>=0.10.0",
         "janis-pipelines.unix>=0.10.0",
     ],
+    extras_require={
+        "ci": [
+            "keyring==21.4.0",
+            "setuptools",
+            "wheel",
+            "twine",
+        ],
+    },
     zip_safe=False,
     long_description=long_description,
     long_description_content_type="text/markdown",