tests.yml

# This workflow is used for testing whether the DEG list produced by the new code is identical to the expected result based on example data.

name: Tests

on:
  push:
    branches:
      - '*'
  
  schedule:
    - cron: '0 0 1 * *'
env:
  CACHE_NUMBER: 0

permissions:
  contents: read

jobs:
  build:
    env:
      GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
    strategy:
      matrix:
        include: 
          - os: ubuntu-20.04
            label: linux-64
            prefix: /usr/share/miniconda3/envs/base
        r-version: ['4.1.1','4.3.1']
        python-version: ['3.7']
        seq_type: ['temposeq'] # add later... , 'rnaseq']
    runs-on: ${{ matrix.os }}
    defaults:
      run:
        shell: bash -l {0}

    steps:
      - uses: actions/checkout@v3
      - name: Install linux dependencies
        run: |
          sudo apt-get update
          sudo apt-get install libcairo2-dev
          sudo apt-get install libxt-dev
      - name: Set up python ${{ matrix.python-version }}
        uses: actions/setup-python@v2
        with:
          python-version: ${{ matrix.python-version }}
      - name: Setup Mambaforge
        uses: conda-incubator/setup-miniconda@v2
        with:
            miniforge-variant: Mambaforge
            miniforge-version: latest
            environment-file: workflow/envs/base.yml
            activate-environment: base
            use-mamba: true
            python-version: ${{ matrix.python-version }}
            mamba-version: "*"
            auto-activate-base: true
            auto-update-conda: true
            use-only-tar-bz2: true # This needs to be set for caching to work properly, according to others
      # - name: Set cache date
      #   run: echo "DATE=$(date +'%Y%m%d')" >> $GITHUB_ENV
      # - uses: actions/cache@v3
      #   with:
      #     path: ${{ matrix.prefix }}
      #     key: ${{ matrix.label }}-conda-${{ hashFiles('workflow/envs/base.yml') }}-${{ env.DATE }}-${{ env.CACHE_NUMBER }}
      #   id: cache
      - name: Update environment
        run: mamba env update -n base -f workflow/envs/base.yml
        if: steps.cache.outputs.cache-hit != 'true'
      - name: Check conda install
        run: |
          conda info
          conda list
        shell: bash -l {0}
      - name: Import data
        run: |
          ls -alht # What is in the "working directory"?
          git clone https://github.com/EHSRB-BSRSE-Bioinformatics/test-data
          rm -r data # Remove existing directory before replacing w/ test data
          rm -r config # Remove existing directory before replacing w/ test data
          mv test-data/${{ matrix.seq_type}}/* ./
          wget https://github.com/EHSRB-BSRSE-Bioinformatics/unify_temposeq_manifests/raw/main/output_manifests/Human_S1500_1.2_standardized.csv
      - name: Build snakemake environment
        run: |
          snakemake --cores 8 --use-conda --conda-create-envs-only
      - name: Install extra dependencies
        run: |
         conda run -p $(grep -rl "R-ODAF_reports" .snakemake/conda/*.yaml | \
         sed s/\.yaml//) Rscript install.R
      - name: Run workflow
        run: |
          snakemake --cores 8 --use-conda
      - name: Generate md5sums
        run: |
          md5sum data/processed/count_table.tsv > checksum.count_table
          md5sum data/metadata/metadata.QC_applied.txt > checksum.metadata.QC_applied
          md5sum analysis/QC/MultiQC_Report_data.zip > checksum.multiQC
          md5sum analysis/QC/details/QC_per_sample.txt > checksum.QC_per_sample
          md5sum analysis/DEG_lists/BaP/*.txt > checksums.BaP_DEGs
          md5sum analysis/DEG_lists/CISP/*.txt > checksums.CISP_DEGs
          # md5sum analysis/pathway_analysis/BaP/*WikiPathways* > checksums.BaP_pathways
          # md5sum analysis/pathway_analysis/CISP/* > checksums.CISP_pathways
          md5sum analysis/BMD_and_biomarker_files/*.txt > checksums.BMD_files
          md5sum analysis/BMD_and_biomarker_files/*/*.txt >> checksums.BMD_files
      - name: Compare pre-processing and QC files to truth set
        run: |
          cmp truth_checksums/checksum.count_table checksum.count_table
          cmp truth_checksums/checksum.metadata.QC_applied checksum.metadata.QC_applied
          #cmp truth_checksums/checksum.multiQC checksum.multiQC
          cmp truth_checksums/checksum.QC_per_sample checksum.QC_per_sample
      - name: Print out checksums for DEG lists
        run: |
          echo "BaP DEG checksums from this run:"
          cat checksums.BaP_DEGs # Can be informative if there are errors
          echo "BaP DEG checksums from 'truth' set:"
          cat truth_checksums/checksums.BaP_DEGs # Can be informative if there are errors
          # There are rounding differences between different environments, e.g.:
          # cat analysis/DEG_lists/BaP/*_significant.txt
          echo "CISP DEG checksums from this run:"
          cat checksums.CISP_DEGs
          echo "CISP DEG checksums from 'truth' set:"
          cat truth_checksums/checksums.CISP_DEGs # Can be informative if there are errors
         #  echo "BaP pathway checksums from this run:"
          # cat checksums.BaP_pathways
          # echo "BaP pathway checksums from 'truth' set:"
          # cat truth_checksums/checksums.BaP_pathways
      - name: Compare DEG lists to truth set
        run: |
          # Only check first column, because of timestamps in filenames
          # cmp <(cat truth_checksums/checksums.BaP_DEGs | awk {'print $1'} | sort) \
          #     <(cat checksums.BaP_DEGs | awk {'print $1'} | sort)
          # cmp <(cat truth_checksums/checksums.CISP_DEGs | awk {'print $1'} | sort) \
          #    <(cat checksums.CISP_DEGs | awk {'print $1'} | sort)
          # Instead of cmp, this code will check if ALL the checksums from the test have a corresponding match in the larger set of possible "truth" checksums.
          # This is a hack, but it should work to deal with rounding issues.
          diff -q <(sort -u checksums.BaP_DEGs | awk {'print $1'}) \
                  <(grep -Fxf \
                  <(cat checksums.BaP_DEGs | awk {'print $1'}) \
                  <(cat truth_checksums/checksums.BaP_DEGs | awk {'print $1'} | sort -u))
          diff -q <(sort -u checksums.CISP_DEGs | awk {'print $1'}) \
                  <(grep -Fxf \
                  <(cat checksums.CISP_DEGs | awk {'print $1'}) \
                  <(cat truth_checksums/checksums.CISP_DEGs | awk {'print $1'} | sort -u))
     # - name: Compare pathway analysis to truth set
        # run: |
          # cmp truth_checksums/checksums.BaP_pathways checksums.BaP_pathways
      - name: Compare BMD input files to truth set
        run: |
          cat checksums.BMD_files # Can be informative if there are errors
          cat truth_checksums/checksums.BMD_files # Can be informative if there are errors
          cmp truth_checksums/checksums.BMD_files checksums.BMD_files