FASTQC problem on read with no sequence data

[turnersd]

I've had a problem with FASTQC. I get the following error:

Failed to process file 0618107SM.fastq
uk.ac.bbsrc.babraham.FastQC.Sequence.SequenceFormatException: Midline 'CAAACATACAGCTTAAAAC
AACAGACATTTATTATCTTATGGT' didn't start with '+'

When I look at the context around CAACATACA..... I see that there is a sequence identifier but no sequence info following it.

@HWUSI-EAS1758R:33:64PA7AAXX:4:1:6103:1039 1:N:0:
CTCGATCCACAAACCGCCCTTGGGGTAAACATTCGG
+
IIIIIIIIIIIIIIIIHIIIIIIIIEHIGHHHHIII
@:701;5677@<5@######################
@HWUSI-EAS1758R:33:64PA7AAXX:4:1:6397:1046 1:N:0:
CCTTAGGTTATTTCATGCCTAGAAATGTATCCTACA
+
HHGHHHGHHDHHHHHH@HHHHHHHHHHGHHHHGHHG
@HWUSI-EAS1758R:33:64PA7AAXX:4:1:6580:1044 1:N:0:
GATCGGAAGAGCTCGTATGCCGTCTTCTGCTTGAAA
+
GGGGEGDDGBHGHHHEGDGEGGD@GBB???GBGDBF

It looks like the identifier that starts as "@:701;5677…" causes a failure at the next read. How would I get rid of these "empty" reads?

[dariober]

How would I get rid of these "empty" reads?

I think the problem is how you got these emtpy reads in the first place!? It looks like something went wrong with some processing upstream. Are you looking at a fastq file straight from the sequencing facility?

Dario

[maubp]

I've had a problem with FASTQC.

No, you've got a problem in your FASTQ file

As Dario says, you'll need to track back and work out what went wrong in the creation of this file. It could be as simple as a data corruption on disk or when copying the file.

[turnersd]

Thanks, yes it's definitely a problem with the FASTQ file. It looks like whoever filtered this data before I got my hands on it was using grep to looks for flags in the identifier and pulls out 4 lines of context around it. But obviously something went very wrong. Now, I just need to hunt down the raw data. Thanks.