replace read end and quality trimming to use fastp

[yannickwurm]

see the title...

[Pigrenok]

But leave cutadapt for cleaning after kmc? I am not sure whether fastp can do the same as the number of reads after kmc is different and fastp does not like it.

[Pigrenok]

There is an incompatibility between fastp and cutadapt. If fastp is used for end and quality trimming using the following commands

fastp -i input/reads.pe1.fastq.gz -o tmp/reads.pe1.trimmed.fq --cut_front 4 --cut_front_mean_quality 10 -h tmp/reads.pe1.fastp.html
fastp -i input/reads.pe2.fastq.gz -o tmp/reads.pe2.trimmed.fq --cut_front 4 --cut_front_mean_quality 10 -h tmp/reads.pe1.fastp.html

then after kmc run, cutadapt gives an error that the paired reads do not match.

So, unless I am using fastp incorrectly, I would say we should stay with cutadapt at least for now.