bcl2fastq produced empty Read 1 (paired end)

Question

[till.huelsmann]

Hello,

I using the bcl2fastq tool and nextseq bcl files.

The bcl files look good but after I use the bcl2fastq tool to produce fastq files (paired end with adapter trimming) the first Read is corrupted and contains no sequences:

size: 545 - NAME_2_S1_R1_001.fastq.gz

size: 5323354262 - NAME_2_S1_R2_001.fastq.gz

Maybe I have to define, that I use paired end in the Sample Sheet?

If I check the archive then I get

NAME_2_S1_R1_001.fastq.gz: ASCII text

NAME_2_S1_R2_001.fastq.gz: gzip compressed data, extra field

But in the logfile (bcl2fastq start with nohup) I can´t see that any error occurs.

And as a result if I try to map it to the reference (using bwa) then I get the following error:

the 1st file has fewer sequences

Hope, that somebody can help me :)

[GenoMax]

Assuming you have done the bcl2fastq demux right there appears to be something wrong with read 1. You can ask Illumina tech support to take a look at this run, if you have never diagnosed these type of issues before.

Post a copy of your RunInfo.xml file (lines that look like <Read Number="1" NumCycles="50" IsIndexedRead="N" />) and SampleSheet.csv, if you are sure the run had no problems.

Answer 1

[till.huelsmann]

It was a bug in the bcl2fastq version 2.19. After I installed the 2.20 version everything is ok.

[GenoMax]

Do you know what this "bug" was and how it was specifically fixed in 2.20?

[till.huelsmann]

Sorry for the late reply.

There is a bug in the bcl2fastq v2.19 version, for some runs it complains about corrupt bcl files which are not corrupt and can cause data to be missing from the FASTQ files. This issue has been resolved in more recent versions of the software such as bcl2fastq v2.19.1 and bcl2fastq v2.20