N base in RNA-seq data

Question

[cellulebioinfobiscem]

Hi, I do an alignement of RNA-seq publish data with STAR aligner. I see in fastqc that for some sample, N content increase to 12% at base 4 and 22 (warning in orange). Should i cut N part ?

Thanks

[GenoMax]

Looks like this data is not good, especially if you consistently have N's at specific cycles. Such data should not have been released by the sequencing facility. You can't remove N's since that will mess up reading frame.

You can filter out reads with N's using reformat.sh from BBMap suite by doing:

reformat.sh in=your_read.fq out=filtered.fq maxns=0

or

reformat.sh in1=your_read.fq in2=your_read.fq out1=filtered.fq out2=filtered.fq maxns=0

(if reads are paired-end).

[cellulebioinfobiscem]

Thanks you for tools suggestion, I will test it.