FastQC Problem

polsum · 11-21-2011, 11:20 AM

Hi while using FastQC, I am getting following error.

uk.ac.bbsrc.babraham.FastQC.Sequence.SequenceFormatException: ID line didn't start with '@'
at uk.ac.bbsrc.babraham.FastQC.Sequence.FastQFile.readNext(FastQFile.java:134)
at uk.ac.bbsrc.babraham.FastQC.Sequence.FastQFile.<init>(FastQFile.java:73)
at uk.ac.bbsrc.babraham.FastQC.Sequence.SequenceFactory.getSequenceFile(SequenceFactory.java:83)
at uk.ac.bbsrc.babraham.FastQC.Analysis.OfflineRunner.processFile(OfflineRunner.java:113)
at uk.ac.bbsrc.babraham.FastQC.Analysis.OfflineRunner.<init>(OfflineRunner.java:84)
at uk.ac.bbsrc.babraham.FastQC.FastQCApplication.main(FastQCApplication.java:308)

THe first line indeed contains some text like file name followed by 00/00/ etc. So I removed these first few lines and tried again. Now it runs fine until 95% analysis completed and gives the same error that the ID doesnt start with @.

THis file is brand new (sequenced last week). But my last year's sequenced file runs just fine with no problems.

Any help is appreciated.

TIA

simonandrews · 11-22-2011, 01:17 AM

It certainly looks like your FastQ file has acquired some extra data it shouldn't have in it which is messing up the parsing. In an older thread I posted a script which you can run which does some sanity checks on FastQ files and writes out a cleaned version of the file which you should then be able to run through FastQC.

polsum · 11-22-2011, 10:35 AM

Quote:

Originally Posted by simonandrews

It certainly looks like your FastQ file has acquired some extra data it shouldn't have in it which is messing up the parsing. In an older thread I posted a script which you can run which does some sanity checks on FastQ files and writes out a cleaned version of the file which you should then be able to run through FastQC.

Thank you very much. This script cleaned my file and now its running just fine.

hartmaier · 06-19-2013, 06:52 AM

Very awesome. The scripts seems to have properly fixed my files and fastqc is now able to run on them.

simonandrews · 06-19-2013, 06:55 AM

Quote:

Originally Posted by hartmaier

Very awesome. The scripts seems to have properly fixed my files and fastqc is now able to run on them.

That's great - but just be aware that something must have corrupted your files in the first place for this script to be needed. This means that you might have lost some unknown amount of data from your files somewhere along the way. If it's possible to go back to the original source of your data then it might be worth seeing if you can find an original un-corrupt version of your data.

hartmaier · 06-19-2013, 07:51 AM

Quote:

Originally Posted by simonandrews

That's great - but just be aware that something must have corrupted your files in the first place for this script to be needed. This means that you might have lost some unknown amount of data from your files somewhere along the way. If it's possible to go back to the original source of your data then it might be worth seeing if you can find an original un-corrupt version of your data.

Yup, thanks. Data was generated elsewhere...will try, but at least I can use the rest of it for now.

deepseq · 02-06-2015, 12:53 PM

How do I run the script on my file on the command line ?? I seem to have a similar issue and would like to use your script....

simonandrews · 02-09-2015, 01:14 AM

You probably shouldn't use that script any more. I wrote it back when Phred64 was being produced by Illumina. Not that everyone's on Phred32 it will give loads of false warnings about quality lines appearing to be headers since an @ symbol can legitimately occur in a quality string in Phred32 (which is why Phred64 was a better idea). It would be possible to make an updated version which was more specific about its checks and would work on Phred32 but I haven't seen one posted.

I should also point out that this is only really a diagnostic to see why a file failed. If it's actually corrupted you've no idea how much data you've lost so this isn't a 'fix' for your data. FastQC also now has more debugging in it to try to tell you why your file failed so just looking at the output from that might be all you need to do.

katka · 11-04-2016, 09:07 AM

Hi,
I had the same problem. As it turned out I used wrong type of decompression of my data(I used bzip2 -dk filename). The file had suffix .bz2 but it was actually double compressed file. So I decompressed it again using tar -xjvf filename and it was no longer a problem.
Maybe, it might be the same case for you and this is the reason why was your fastq file corrupted.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Where is FastQC?	sklages	General	10	02-06-2012 12:46 AM
fastQC	papori	RNA Sequencing	3	02-04-2012 02:48 PM
Need help for FastQC results. Thanks!!	byou678	Bioinformatics	18	08-23-2011 02:53 PM
fastqc with 454?	newbie25	Bioinformatics	3	07-07-2011 12:52 PM
Fastqc error	Seq84	Bioinformatics	0	04-27-2011 08:22 AM

11-21-2011, 11:20 AM	#1
polsum Member Location: Texas Join Date: May 2009 Posts: 32	FastQC Problem Hi while using FastQC, I am getting following error. uk.ac.bbsrc.babraham.FastQC.Sequence.SequenceFormatException: ID line didn't start with '@' at uk.ac.bbsrc.babraham.FastQC.Sequence.FastQFile.readNext(FastQFile.java:134) at uk.ac.bbsrc.babraham.FastQC.Sequence.FastQFile.<init>(FastQFile.java:73) at uk.ac.bbsrc.babraham.FastQC.Sequence.SequenceFactory.getSequenceFile(SequenceFactory.java:83) at uk.ac.bbsrc.babraham.FastQC.Analysis.OfflineRunner.processFile(OfflineRunner.java:113) at uk.ac.bbsrc.babraham.FastQC.Analysis.OfflineRunner.<init>(OfflineRunner.java:84) at uk.ac.bbsrc.babraham.FastQC.FastQCApplication.main(FastQCApplication.java:308) THe first line indeed contains some text like file name followed by 00/00/ etc. So I removed these first few lines and tried again. Now it runs fine until 95% analysis completed and gives the same error that the ID doesnt start with @. THis file is brand new (sequenced last week). But my last year's sequenced file runs just fine with no problems. Any help is appreciated. TIA

02-06-2015, 12:53 PM	#7
deepseq Junior Member Location: california Join Date: Apr 2011 Posts: 2	same fastq error How do I run the script on my file on the command line ?? I seem to have a similar issue and would like to use your script....

11-04-2016, 09:07 AM	#9
katka Junior Member Location: Brno Join Date: Nov 2016 Posts: 1	corrupted file Hi, I had the same problem. As it turned out I used wrong type of decompression of my data(I used bzip2 -dk filename). The file had suffix .bz2 but it was actually double compressed file. So I decompressed it again using tar -xjvf filename and it was no longer a problem. Maybe, it might be the same case for you and this is the reason why was your fastq file corrupted.

11-22-2011, 01:17 AM	#2
simonandrews Simon Andrews Location: Babraham Inst, Cambridge, UK Join Date: May 2009 Posts: 871	It certainly looks like your FastQ file has acquired some extra data it shouldn't have in it which is messing up the parsing. In an older thread I posted a script which you can run which does some sanity checks on FastQ files and writes out a cleaned version of the file which you should then be able to run through FastQC.

06-19-2013, 06:52 AM	#4
hartmaier Member Location: Pittsburgh Join Date: Dec 2012 Posts: 12	Very awesome. The scripts seems to have properly fixed my files and fastqc is now able to run on them.

02-09-2015, 01:14 AM	#8
simonandrews Simon Andrews Location: Babraham Inst, Cambridge, UK Join Date: May 2009 Posts: 871	You probably shouldn't use that script any more. I wrote it back when Phred64 was being produced by Illumina. Not that everyone's on Phred32 it will give loads of false warnings about quality lines appearing to be headers since an @ symbol can legitimately occur in a quality string in Phred32 (which is why Phred64 was a better idea). It would be possible to make an updated version which was more specific about its checks and would work on Phred32 but I haven't seen one posted. I should also point out that this is only really a diagnostic to see why a file failed. If it's actually corrupted you've no idea how much data you've lost so this isn't a 'fix' for your data. FastQC also now has more debugging in it to try to tell you why your file failed so just looking at the output from that might be all you need to do.