Short reads pre-processing before polishing

[aureliendejode]

Question or Expected behavior
I have a reference genome obtained after CANU assembly, purge_dups and long reads polishing with arrow. I'd like to polish with short reads using Next_polish.
Should I use raw Illumina reads or do some pre-processing (e.g. reads trimming, remove low quality reads ...) of the short reads before using them for polishing ?

Thank you

[moold]

Yes

[aureliendejode]

Ok good.
My next question is there a recommend depth of coverage for polishing ?
More precisely a maximum depth of coverage because I have close to 500X of Illumina raw reads, but I guess there is no need to use all of it ?

[moold]

I do not test, but you can try to set -max_depth to control haw many reads are used to polish, see here