Check and test NCBI updates #84
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: Check and test NCBI updates | |
| on: | |
| workflow_dispatch: | |
| pull_request: # tests whether it is working on PR | |
| paths: | |
| - '.github/workflows/ncbi.yml' | |
| schedule: | |
| - cron: "0 0 1,15 * *" # Run the workflow on the 1st and 15th day of each month | |
| permissions: | |
| contents: write | |
| pages: write | |
| id-token: write | |
| issues: write | |
| jobs: | |
| check_new_data: | |
| runs-on: ubuntu-latest | |
| name: Check the date of the latest data | |
| outputs: | |
| DATE_NEW: ${{ steps.check_download.outputs.DATE_NEW }} | |
| DATE_OLD: ${{ steps.check_download.outputs.DATE_OLD }} | |
| steps: | |
| # step 1: check the release date for the latest NCBI files | |
| - name: Checkout | |
| uses: actions/checkout@v4 | |
| - name: Check for new ncbi files | |
| id: check_download | |
| run: | | |
| ##Extract the date from the ncbi config file | |
| date_old=$(grep -E '^date=' datasources/ncbi/config | cut -d'=' -f2) | |
| echo 'Accessing the ncbi data' | |
| last_modified=$(curl -sI https://ftp.ncbi.nih.gov/gene/DATA/gene_history.gz | grep -i Last-Modified) | |
| ##Extract the date from the latest changes (up to the day) | |
| date_new=$(echo $last_modified | cut -d':' -f2- | xargs -I {} date -d "{}" +%Y-%m-%d) | |
| #Store dates to output | |
| echo "DATE_OLD=$date_old" >> $GITHUB_OUTPUT | |
| echo "DATE_NEW=$date_new" >> $GITHUB_OUTPUT | |
| echo "Date of latest release: $date_new", "Date of release of the current version: $date_old" | |
| test_new_data_processing: | |
| name: Processing new data and check updates | |
| needs: check_new_data | |
| env: | |
| DATE_OLD: ${{ needs.check_new_data.outputs.DATE_OLD }} | |
| DATE_NEW: ${{ needs.check_new_data.outputs.DATE_NEW }} | |
| runs-on: ubuntu-latest | |
| steps: | |
| - name: Checkout repository | |
| uses: actions/checkout@v4 | |
| # step 2: download the recent data | |
| - name: Download the recent data | |
| run: | | |
| ##Store outputs from previous job in environment variables | |
| echo "$DATE_NEW=$DATE_NEW" >> $GITHUB_ENV | |
| ##Create temp. folder to store the data in | |
| mkdir -p datasources/ncbi/data | |
| ##Download ncbi file | |
| wget https://ftp.ncbi.nih.gov/gene/DATA/gene_info.gz | |
| wget https://ftp.ncbi.nih.gov/gene/DATA/gene_history.gz | |
| mv gene_info.gz gene_history.gz datasources/ncbi/data | |
| ##Check file size if available | |
| ls -trlh datasources/ncbi/data | |
| # step 3: run the Rscripts for ncbi preprocessing | |
| - name: Install R | |
| uses: r-lib/actions/setup-r@v2 | |
| with: | |
| r-version: '4.1' # Specify the R version | |
| include-recommended: true | |
| - name: Test ncbi data processing | |
| id: ncbi_process | |
| run: | | |
| sourceVersion=$DATE_NEW | |
| gene_history="data/gene_history.gz" | |
| gene_info="data/gene_info.gz" | |
| # Run rscripts program and capture its exit code | |
| Rscript r/src/ncbi.R $sourceVersion $gene_history $gene_info | |
| # Check the exit status of the R script | |
| if [ $? -eq 0 ]; then | |
| # script succeeded | |
| echo "Successful preprocessing of ncbi data." | |
| echo "FAILED=false" >> $GITHUB_ENV | |
| else | |
| # script failed | |
| echo "Failed preprocessing of ncbi data." | |
| echo "FAILED=true" >> $GITHUB_ENV | |
| fi | |
| # step 4: compare the new and old data | |
| - name: RegEx and Diff test | |
| if: | |
| ${{ env.FAILED == 'false' }} | |
| run: | | |
| # Set up vars from config file | |
| to_check_from_zenodo=$(grep -E '^to_check_from_zenodo=' datasources/ncbi/config | cut -d'=' -f2) | |
| old="datasources/ncbi/data/$to_check_from_zenodo" | |
| new="datasources/ncbi/recentData/$to_check_from_zenodo" | |
| column_name="secondaryID" | |
| echo $column_name | |
| # Unzip the zip file | |
| unzip datasources/ncbi/data/NCBI_secID2priID.zip -d datasources/ncbi/data/ | |
| # remove headers | |
| sed -i '1d' "$new" | |
| sed -i '1d' "$old" | |
| # qc integrity of IDs | |
| wget -nc https://raw.githubusercontent.com/bridgedb/datasources/main/datasources.tsv | |
| NCBI_ID=$(awk -F '\t' '$1 == "Entrez Gene" {print $10}' datasources.tsv) | |
| # Split the file into two separate files for each column | |
| awk -F '\t' '{print $1}' $new > column1.txt | |
| awk -F '\t' '{print $2}' $new > column2.txt | |
| # Use grep to check if any line in the primary column doesn't match the pattern | |
| if grep -nqvE "$NCBI_ID" "column1.txt"; then | |
| echo "All lines in the primary column match the pattern." | |
| else | |
| echo "Error: At least one line in the primary column does not match pattern." | |
| grep -nvE "^$NCBI_ID$" "column1.txt" | |
| echo "FAILED=true" >> $GITHUB_ENV | |
| exit 1 | |
| fi | |
| # Use grep to check if any line in the secondary column doesn't match the pattern | |
| if grep -nqvE "$NCBI_ID" "column1.txt"; then | |
| echo "All lines in the secondary column match the pattern." | |
| else | |
| echo "Error: At least one line in the secondary column does not match pattern." | |
| grep -nqvE "$NCBI_ID" "column2.txt" | |
| echo "FAILED=true" >> $GITHUB_ENV | |
| exit 1 | |
| fi | |
| # sort them | |
| cat "$old" | sort | tr -d "\r" | cut -f 1,3 > ids_old.txt | |
| cat "$new" | sort | tr -d "\r" | cut -f 1,3 > ids_new.txt | |
| # TODO decide whether to perform diff on symbols too? | |
| ##cat "$old" | sort | tr -d "\r" | cut -f 4 > symbols_old.txt | |
| ##cat "$new" | sort | tr -d "\r" | cut -f 4 > symbols_new.txt | |
| echo "Performing diff between the sorted lists of IDs" | |
| # Perform a diff between the sorted lists of IDs | |
| output_file=diff.txt | |
| diff -u ids_old.txt ids_new.txt > $output_file || true | |
| # retrieve new lines | |
| added=$(grep '^+' "$output_file" | sed 's/+//g') || true | |
| # retrieve removed lines | |
| removed=$(grep '^-' "$output_file" | sed 's/-//g') || true | |
| added_filtered=$(comm -23 <(sort <<< "$added") <(sort <<< "$removed")) | |
| removed_filtered=$(comm -23 <(sort <<< "$removed") <(sort <<< "$added")) | |
| added=$added_filtered | |
| removed=$removed_filtered | |
| # count them | |
| count_removed=$(printf "$removed" | wc -l) || true | |
| count_added=$(printf "$added" | wc -l) || true | |
| # make sure we are not counting empty lines | |
| if [ -z "$removed" ]; then | |
| count_removed=0 | |
| removed="None" | |
| fi | |
| if [ -z "$added" ]; then | |
| count_added=0 | |
| added="None" | |
| fi | |
| echo ________________________________________________ | |
| echo " removed pairs " | |
| echo ________________________________________________ | |
| echo "$removed" | |
| echo ________________________________________________ | |
| echo " added pairs " | |
| echo ________________________________________________ | |
| echo "$added" | |
| echo _________________________________________________ | |
| echo "What's changed:" | |
| echo "- Added id pairs: $count_added" | |
| echo "- Removed id pairs: $count_removed" | |
| # Store to env to use in issue | |
| echo "ADDED=$count_added" >> $GITHUB_ENV | |
| echo "REMOVED=$count_removed" >> $GITHUB_ENV | |
| count=$(expr $count_added + $count_removed) || true | |
| echo "COUNT=$count" >> $GITHUB_ENV | |
| total_old=$(cat "$old" | wc -l) || true | |
| change=$((100 * count / total_old)) | |
| echo "CHANGE=$change" >> $GITHUB_ENV | |
| - name: 'Upload processed data as artifacts' | |
| uses: actions/upload-artifact@v3 | |
| with: | |
| name: ncbi_processed | |
| path: datasources/ncbi/recentData/* | |
| - uses: JasonEtco/create-an-issue@v2 | |
| if: env.COUNT != 0 | |
| name: Post issue about update availability | |
| env: | |
| GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }} | |
| SOURCE: "NCBI" | |
| with: | |
| filename: .github/ISSUE_TEMPLATE/ISSUE_UPDATE.md | |
| update_existing: true | |
| - uses: JasonEtco/create-an-issue@v2 | |
| name: Post issue about failing test | |
| if: ${{ env.FAILED == 'true' }} | |
| env: | |
| GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }} | |
| SOURCE: ncbi | |
| with: | |
| filename: .github/ISSUE_TEMPLATE/ISSUE_FAIL.md | |
| update_existing: true |