How filter bcl2fastq reads ?

[ClemBuntu]

Hello everyone,

I got a nextseq run that I demultiplexed using bcl2fastq.
Then I used fastq illumina filter and I got this strange result :

fastq_illumina_filter (--keep N) statistics:
Input: 141,900,316 reads
Output: 141,900,316 reads (9%)

According to the website where I've downloaded this software :

This program can filter FASTQ files produced by CASAVA 1.8

Indeed, all my reads (the 4 lanes) have the "N" tag I checked with the following command

grep -A 3 '^@.* [^:]*:Y:[^:]*:'

Why all my reads have the "N" tag ? According to the bcl2fastq user guide this tag DOES exist with bcl2fastq ! (page 22) So, is there a way to filter reads which are produced with bcl2fastq ? Do I have to update the RTA version ? And finally why fastq_illumina_filter gave me 9% of read passing filter ?

[GenoMax]

Forgive me if I am not understanding your question.

Only way to get "failed" reads in output file is to run bcl2fastq with the option "--with-failed-reads".

By default N stands for NOT filtered i.e. of good quality.

[ClemBuntu]

Only way to get "failed" reads in output file is to run bcl2fastq with the option "--with-failed-reads".

That's why I wasn't able to see them, thanks