FastQC Report: A horn in Per sequence GC content ?!

[leekb]

Hi Everyone,

Nice to meet you guys here!
I'm doing ChIP-Seq using HiSeq 1500 on SR 50 cycles, multiplexed condition.
And I got a funny curve in my FastQC report on Per sequence GC content and Sequence Duplication Levels as well as overrepresented sequence.
Anyone please give a help
1) Why my Per sequence GC content curve got a horn?
2) How come such high sequence duplication level like 71.09% ? Due to overrepresented sequence?
3) As attached BioAnalyzer, there's totally absence of adapter, why still overrepresented sequence like Index 7?

Library prep condition: 12cycles PCR amplification

Cheers!
kb

[leekb]

One more strange FastQC report

Hi Everyone,

Nice to meet you guys here!
I'm doing ChIP-Seq using HiSeq 1500 on SR 50 cycles, multiplexed condition.
And I got a funny curve in my FastQC report on Per sequence GC content and Sequence Duplication Levels as well as overrepresented sequence.
Anyone please give a help
1) Why my Per sequence GC content curve got a horn?
2) How come such high sequence duplication level like 71.09% ? Due to overrepresented sequence?
3) As attached BioAnalyzer, there's totally absence of adapter, why still overrepresented sequence like Index 7?

Library prep condition: 12cycles PCR amplification

Cheers!
kb

Hi all,

More strange FastQC report.
This's time I got problems on
1)Per base sequence quality: why quality drop dramatically in base 23-25? Is there a bubble inside the lane?
2)per base N content : a small peak shows up in base 23, any relationship with quality drop in per base sequence result?
3)sequence duplication levels : a little bit difference from above one, but more serious in duplication.

In general, is't a fair good result or in very pool quality? Should I keep continue the analysis?

kb

[GenoMax]

Hi all,

More strange FastQC report.
This's time I got problems on
1)Per base sequence quality: why quality drop dramatically in base 23-25? Is there a bubble inside the lane?
2)per base N content : a small peak shows up in base 23, any relationship with quality drop in per base sequence result?
3)sequence duplication levels : a little bit difference from above one, but more serious in duplication.

In general, is't a fair good result or in very pool quality? Should I keep continue the analysis?

kb

It is possible that there was a bubble in the lane (or some other issue) around cycle 23. It does not appear to be a severe problem though since most of bases are still appear to be of good quality. Hopefully during alignment these reads would not align and will be discarded.

Since you are doing ChIP-seq you may be enriching a binding sequence that may have a GC predominance. You should continue with the analysis.

Just to be safe run your sequences through "trimmomatic" (http://www.usadellab.org/cms/?page=trimmomatic) to make sure that there are no primer dimer and other odd contaminants.

[leekb]

It is possible that there was a bubble in the lane (or some other issue) around cycle 23. It does not appear to be a severe problem though since most of bases are still appear to be of good quality. Hopefully during alignment these reads would not align and will be discarded.

Since you are doing ChIP-seq you may be enriching a binding sequence that may have a GC predominance. You should continue with the analysis.

Just to be safe run your sequences through "trimmomatic" (http://www.usadellab.org/cms/?page=trimmomatic) to make sure that there are no primer dimer and other odd contaminants.

Thanks GenoMax!

Yes you're right, though cycle 23-25 drop quality but still stays in green zone.

How about my duplication levels issue? It's crazy high and looks like not much unique reads left for me to analysis. Any way to improve next time?
Many thanks!

kb