Qualimap module missing "percentage_aligned" after 1.16

[Stikus]

Description of bug

After last Qualimap module update #2020 behavior changed, and we lost "percentage_aligned" sections in our reports.

Looks like the reason behind it is moving percentage_aligned into additional try:

Our report doesn't have Globals inside section (due to full genome analysis) and looks like try causing missing entire block - not only percentage_aligned_on_target section.

I'm not using last version (due to Python 3.8 requirement), but according to blame and changelogs - nothing is changed in this part.

File that triggers the error

genome_results.txt

MultiQC Error log

$ LC_ALL=C.UTF-8 LANG=C.UTF-8 /soft/MultiQC-1.16/bin/multiqc --fullnames --outdir /outputs --filename test_full_filtered_multiqc_report.html /outputs/inputMultiQC

  /// MultiQC 🔍 | v1.16

|           multiqc | MultiQC Version v1.17 now available!
|           multiqc | Not cleaning sample names
|           multiqc | Search path : /outputs/inputMultiQC
|         searching | ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% 5/5  
|            picard | Found 1 MarkDuplicates reports
|          samtools | Found 2 flagstat reports
|            fastqc | Found 2 reports
|           multiqc | Report      : test_full_filtered_multiqc_report.html
|           multiqc | Data        : test_full_filtered_multiqc_report_data
|           multiqc | MultiQC complete

Before submitting

• I have read the troubleshooting documentation.
• I am using the latest release of MultiQC.
• I have included a full MultiQC log, not truncated.
• I have attached an input file (.zip if necessary) that triggers the error.

[vladsavelyev]

Thanks so much for reporting the issue, and providing a test example!
Should be fixed in #2207 😌
I added you example in the test data repo: MultiQC/test-data#308

[Stikus]

@vladsavelyev Hello, we rechecked results after 1.19 release - looks like there is one major difference in report_saved_raw_data section:

general_error_rate changed from 0.0068 to 0.6799999999999999 (100 time more) likely due to 100 multiplier in code. It this intended? In original file (attached in issue) we have:

>>>>>>> Mismatches and indels

    general error rate = 0.0068
    number of mismatches = 48,036
    number of insertions = 648
    mapped reads with insertion percentage = 0.78%
    number of deletions = 734
    mapped reads with deletion percentage = 0.89%
    homopolymer indels = 57.96%