Extension: Quality Settings

Data and software

This extension follows on from the tutorial on primer removal using the same data and software.

Because primers are a region on our sequence that we have some a priori knowledge about, this is a good opportunity for filtering sequences with errors.

Exercise

• Try running primer trimming with the strictest settings, a 100% match in length and bases
• Is this sensible?

Many metabarcoding pipelines trim primers by just trimming a number of bases equal to the primer length off the beginning of each read. You can do this in cutadapt using the ​``-u`` ​option, which you would need to do for each direction separately (replacing N with the correct primer length:

cutadapt -u N -o out.fastq -i in.fastq

Exercise

• Try running this command for the forward and reverse reads of sample.
• If you look at the number of reads, this clearly retains more.
• What might be the downsides of doing it this way?