NextSeq auto-detect seems incomplete

[claczny]

Hi,

first of all, thank you for developing the software!

I ran it on some samples that were sequenced with a NextSeq 2000. For some samples, the reads are mostly composed of "G", which is something we need to look into, so this is not the issue :)

The issue is that the autodetection seems to fail, hence the --trim_poly_g needs to be specified manually as it is otherwise not enabled. When I look at https://github.com/OpenGene/fastp/blob/ca559a71feed94e74ea449e7567d0506de48dea4/src/evaluator.cpp#L25, the prefixes that are used to identify the machine are very limited. Specifically, the FASTQ files, I have, start with @VH for the readnames. That prefix is, however, not in the list. Maybe because these prefixes were for NextSeq 5xx and not NextSeq 2000?

Best wishes and stay safe,

Cedric