How not to merge R1 and R2 pair-end files in multiqc report

Good morning,

Could you please advise how to generate multiqc report without merging R1 and R2 pair-end files?

My files look like H_SAMPLE2_S29.trim_fin.R1_fastqc.zip, H_SAMPLE2_S29.trim_fin.R2_fastqc.zip. Multiqc generates H_SAMPLE2_S29 stats in the report. But I expect to have stats for R1 and R2 separately.

v1.12 and v1.14 versions give the same result. I have tried --file-list, but it didn't help. I also tried to rename zips, but, as I understand it, renaming the contents of the files in zips is required. Maybe there is an easy way to set a specific multiqc flag? --fn_as_s_name doesn't lead to the desired result.

Thank you in advance!

Best regards,
Poecile

multiqc fastqc

• 2,924 views • link

written 3.5 years ago by poecile.pal •

Hi Poecile,

Isn't that what multiqc is supposed to do? Maybe you should run fastQC to get separate results.

• 0 views • link

written 3.5 years ago by SushiRoll •

SushiRoll,

Thanks for the answer!

I use multiqc to combine the results obtained by fastqc (yes, fastqc reports folder contains reports for R1 and R2 separately: H_SAMPLE2.trim_fin.R1_fastqc.zip, H_SAMPLE2.trim_fin.R2_fastqc.zip).

multiqc generates the report with tables and graphs in which the individual units are rows like H_SAMPLE1, H_SAMPLE2. For example, %GC for H_SAMPLE1, %GC for H_SAMPLE2. I would like to see %GC separately for H_SAMPLE 1.R1 and H_SAMPLE1.R1, H_SAMPLE2.R1 and H_SAMPLE2.R2 in the final table. But multiqc automatically combines paired samples.

• 0 views • link

written 3.5 years ago by poecile.pal •

1 answer

It's because you have _trim in the filenames before the _R1 bit. MultiQC is "cleaning" the sample names back to this, which makes them identical and overwrite one another.

To fix this you can:

Change the filenames to move the _R1 bit earlier in the filename (then rerun FastQC and MultiQC)
Stop MultiQC from cleaning any sample names with the --fullnames flag (see docs)
Customise MultiQC cleaning patterns to remove the _trim string (see docs)

• 0 views • link

written 3.5 years ago by Phil Ewels

Phil Ewels,

Thank you so much for your help!

It was too long for me to rerun fastqc due to the large number of samples, so, I renamed all archives and then their contents:

unzip SAMPLE11_R1_001_trim_fin_fastqc.zip && cd SAMPLE11.trim_fin.R1_fastqc && sed -i 's/\.trim_fin\.R1\.fastq\.gz/_R1_001_trim_fin_fastqc\.gz/' summary.txt && sed -i 's/\.trim_fin\.R1\.fastq\.gz/_R1_001_trim_fin_fastqc\.gz/' fastqc_data.txt && sed -i 's/\.trim_fin\.R1\.fastq\.gz/_R1_001_trim_fin_fastqc\.gz/' fastqc_report.html && cd .. && rm SAMPLE11_R1_001_trim_fin_fastqc.zip && zip -r SAMPLE11_R1_001_trim_fin_fastqc.zip SAMPLE11.trim_fin.R1_fastqc && rm -r SAMPLE11.trim_fin.R1_fastqc

(and similarly for other files)

Indeed, multiqc separated paired samples.

But, of course, it's not very convenient) So I have tried using the flag --fullnames, and it works!

Thanks again!

• 0 views • link

written 3.5 years ago by poecile.pal •