java.lang.IndexOutOfBoundsException

[Koen Nijbroek]

Hi All,

I have processed a 120 BAM files in parallel (cluster) via bamqc, which worked perfectly fine! Now I'm trying to perform a multi-bamqc on those 120 BAM files, and that gives me errors. I have already browsed through some dicussions here with similar errors, but that didn't help in solving my problem. Unfortunately I cannot share any of my data.....

QualiMap v.2.1
Built on 2015-03-19 12:05
Selected tool: multi-bamqc
Tue May 19 09:02:04 CEST 2015           WARNING output folder already exists
Running multi-sample BAM QC
Checking input paths
Loading sample data
Creating charts
Failed to run multi-bamqc
java.lang.IndexOutOfBoundsException: Index: 0, Size: 0
        at java.util.ArrayList.rangeCheck(ArrayList.java:635)
        at java.util.ArrayList.get(ArrayList.java:411)
        at org.bioinfo.ngs.qc.qualimap.beans.XYVector.get(XYVector.java:104)
        at org.bioinfo.ngs.qc.qualimap.beans.BamQCChart.render(BamQCChart.java:270)
        at org.bioinfo.ngs.qc.qualimap.process.MultisampleBamQcAnalysis.createHistogramBasedChart(MultisampleBamQcAnalysis.java:289)
        at org.bioinfo.ngs.qc.qualimap.process.MultisampleBamQcAnalysis.createCharts(MultisampleBamQcAnalysis.java:500)
        at org.bioinfo.ngs.qc.qualimap.process.MultisampleBamQcAnalysis.run(MultisampleBamQcAnalysis.java:86)
        at org.bioinfo.ngs.qc.qualimap.main.MultisampleBamQcTool.execute(MultisampleBamQcTool.java:157)
        at org.bioinfo.ngs.qc.qualimap.main.NgsSmartTool.run(NgsSmartTool.java:187)
        at org.bioinfo.ngs.qc.qualimap.main.NgsSmartMain.main(NgsSmartMain.java:103)
Tue May 19 10:34:58 CEST 2015           WARNING Cleanup output dir

I hope you can help me!

Thanks in advance,

Koen

[Koen Nijbroek]

PS:

[kni@gyges Mapping]$ qualimap multi-bamqc -d qualimap.list -outdir multi-bamqc --java-mem-size=800G
[kni@gyges Mapping]$ head qualimap.list
BRA_0065        BRA_0065.bam.qualimap
BRA_0381        BRA_0381.bam.qualimap
BRA_0390        BRA_0390.bam.qualimap
BRA_0393        BRA_0393.bam.qualimap
BRA_0399        BRA_0399.bam.qualimap
BRA_0402        BRA_0402.bam.qualimap

[Vang Le Quy]

Hi Koen

It is not necessary you share your whole data or a significant part of it. You only need to produce a subsample of one bam file and see if you can reproduce the problem with that small BAM file.

For example: 

samtools view -s 0.01 -h -b big.bam > small.sam

samtools view -h -bS small.sam > small.bam

You can edit the small.sam like delete sample name, or any sensitive information before sending out. 

I think that will help others and the developers to find out what's wrong, while keeping your data safe.

Vang

On Tuesday, May 19, 2015 at 11:15:18 AM UTC+2, Koen Nijbroek wrote:

[Koen Nijbroek]

Thanks for that tip.. Will take it in consideration.

Furthermore I've (multi) processed 12 BAM files, same method, other genome / other project - which works perfectly.... !

[Koen Nijbroek]

I have done exactly the same on a small subset of 2/120 problematic 'big' BAM files... and that works. Anyone who can help me?

[Konstantin Okonechnikov]

Hi Koen!

Thanks a lot for report! 

As I understood BAM QC run normally on each file, however Multi BAM QC failed when processing the results. 

 

Could you please share all BAM QC results, so I can check this issue carefully? For example, copy all BAM QC result folders in one folder and pack it.

Additionally I will try to create 120 random subsamples and run this too (perhaps the reason of fail is exactly the number samples)

--

  Konstantin

On Tue, May 26, 2015 at 8:39 PM, Koen Nijbroek <koenni...@gmail.com> wrote:

I have done exactly the same on a small subset of 2/120 problematic 'big' BAM files... and that works. Anyone who can help me?

--
You received this message because you are subscribed to the Google Groups "QualiMap" group.
To unsubscribe from this group and stop receiving emails from it, send an email to qualimap+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Koen Nijbroek]

i'll do that asap. I've got the feeling that it has to do with the amount as well... a subset of 20 for example also works fine.

Op woensdag 27 mei 2015 17:15:56 UTC+2 schreef Konstantin Okonechnikov:

[Konstantin Okonechnikov]

Hi Koen,

i wrote a BASH script to create 120 subsamples from a BAM file, then launch qualimap BAM QC on each of them and also generate Multi-sample BAM QC config file.

The script is attached.

Then I tried running it. After the script finished, I run Multi-sample BAM QC and it worked fine on 120 BAM QC outputs. The results are also attached.

Since I did not detect any bug, it would be great if you share exactly your Qualimap BAM QC result folders, so I will be able to check them carefully. Such archive should not be too big.

Let me know if your have any questions.

--

  Konstantin