This is a test version of Biostars. For the public version, visit https://www.biostars.org.
why my fastqc on trimmed data failed?

Hi

I prepared a FastQC report from my raw FASTQ files before trimming. The results showed:

Per tile sequence quality: Failed
Per base sequence content: Failed
Sequence Duplication Levels: Warning (yellow flag) with "percentage of sequences remaining if duplicated: 55.16%".

I then performed trimming using Trimmomatic. After trimming, I ran FastQC again, and the report shows:

Per tile sequence quality: Failed
Per base sequence content: Failed
Sequence Duplication Levels: Warning (yellow flag) with "percentage of sequences remaining if duplicated: 64.94%"
Sequence Length Distribution: Warning (yellow flag)

My questions are:

  1. Why do Per tile sequence quality and Per base sequence content still fail after trimming? Is this expected, or does it indicate a sequencing problem?
  2. The Sequence Duplication Levels warning actually increased after trimming or should I be concerned?
  3. Does the yellow flag in Sequence Length Distribution indicate a problem, or is it normal after trimming?
  4. How should I interpret these FastQC warnings for downstream RNA-seq analysis , can I proceed, or do I need to re-sequence or perform additional filtering?
fastqc galaxy analyses ngs
• 481 views • link
updated by Ram Ram • written by kamran.kouchakzadeh.kk • kamran.kouchakzadeh.kk

2 answers

There are multiple blog posts regarding fastqc metrics interpretation by the authors at: https://sequencing.qcfail.com/software/fastqc/

Specifically check the following:
https://sequencing.qcfail.com/articles/libraries-can-contain-technical-duplication/
https://sequencing.qcfail.com/articles/positional-sequence-bias-in-random-primed-libraries/
It may be best to provide/attach screenshots of plots you are referring.

• 0 views • link
written by GenoMax GenoMax

You should also consult the FastQC documentation where each of the analysis modules are documented:

https://www.bioinformatics.babraham.ac.uk/projects/fastqc/Help/3%20Analysis%20Modules/

For example, the "Per Tile Sequencing Quality":

https://www.bioinformatics.babraham.ac.uk/projects/fastqc/Help/3%20Analysis%20Modules/12%20Per%20Tile%20Sequence%20Quality.html

It will help you put your data into better context:

Per Tile Sequence Quality: This module will issue a warning if any tile shows a mean Phred score more than 5 less than the mean for that base across all tiles.

Though, now that I read that manual myself, it appears to be less than clear. First it has a typo, it says issue a warning when it should say issue a failure in addition the verbiage seems pretty confusing : shows a mean Phred score more than 5 less than the mean ... I wouldn't blame anyone who does not immediatly understand what that means

But long story short, always make sure you understand why a failure occures, what characteristics triggered the fail signal.

• 0 views • link
written by Istvan Albert Istvan Albert

Log in to answer this question.

More posts like this