Still bad phred score after fastp

Question

[SpliceAndScript]

I have performed fastp on my files, trimmed the adaptor sequences, and raw data quality has become better. But in all of my R1 reads I am observing a certain pattern in per base sequence quality in which box-and-whisker plot is going in red region at 3 base positions. In all of my R2 reads poly-G adaptor sequence is removed, but still per base sequence quality is very bad. At a lot of base positions the plot is going in red region. What should i do now, and how much should i trim it again?

[GenoMax]

If you can include screenshots of what you are referring to in text that would help get specific comments. In absence of the images, it sounds like there was some issue with this run at the end of read 1. If that is the case no amount of trimming is likely to fix the issue with phred scores. If you have a good reference you can align to then you may be able to overlook the bad scores and still get some usable data.

There are some good blog posts by authors of FastQC here: https://sequencing.qcfail.com/software/fastqc/ See if anything in there helps.

[AIMAR]

Hi I didn't really understand what's the problem, so if you can provide some images also the FastP command that you've used it could be helpfull to understand the problem and try to find a solution.

thanks !

[Arup Ghosh]

By default, the quality score threshold is 15. So if you have not specified a value for --qualified_quality_phred or --average_qual, there will be bases in the red region.