High percentage of N calls in the library

apredeus · 12-08-2012, 08:35 AM

hello all,

I've discovered an peculiar feature in the libraries that I'm working with using FastQC

The percentage of "N" calls is dramatically higher for base #11. It's at ~ 10% level (as opposed to << 1 % for all other positions). This persists across several libraries and it's always the base #11. The reads are 42 nt long.

Does this mean there was some problem with the instrument? Can't imagine it would be something in the library construction...

Otherwise base call Phred scores are in good to excellent range. The library is also very repetitive.

MeganS · 12-08-2012, 11:53 AM

I think this is likely an issue during sequencing rather than library prep. We see a similar thing occasionally and when it occurs it is always seen in all samples that were run together in a lane, but not in other samples that had libraries prepared at the same time.

apredeus · 12-08-2012, 04:27 PM

Thank you. That's kind of what I thought.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Questions on High% Dot+Mixed and library cpb	WSL	454 Pyrosequencing	1	10-05-2012 01:44 AM
Homozygous Ref calls and no calls	ashkot	Bioinformatics	1	08-09-2012 11:01 AM
Indexed Mate Pair Library Prep for GAIIx/High Seq	busypops	Illumina/Solexa	3	03-10-2012 01:46 AM
Same library to get different percentage of enriched beads	g861@hotmail.com	454 Pyrosequencing	17	08-26-2011 11:05 AM
PubMed: Large scale library generation for high throughput sequencing.	Newsbot!	Literature Watch	0	05-20-2011 12:30 AM

12-08-2012, 08:35 AM	#1
apredeus Senior Member Location: Bioinformatics Institute, SPb Join Date: Jul 2012 Posts: 151	High percentage of N calls in the library hello all, I've discovered an peculiar feature in the libraries that I'm working with using FastQC The percentage of "N" calls is dramatically higher for base #11. It's at ~ 10% level (as opposed to << 1 % for all other positions). This persists across several libraries and it's always the base #11. The reads are 42 nt long. Does this mean there was some problem with the instrument? Can't imagine it would be something in the library construction... Otherwise base call Phred scores are in good to excellent range. The library is also very repetitive.

12-08-2012, 11:53 AM	#2
MeganS Member Location: US Join Date: Sep 2010 Posts: 14	I think this is likely an issue during sequencing rather than library prep. We see a similar thing occasionally and when it occurs it is always seen in all samples that were run together in a lane, but not in other samples that had libraries prepared at the same time.

12-08-2012, 04:27 PM	#3
apredeus Senior Member Location: Bioinformatics Institute, SPb Join Date: Jul 2012 Posts: 151	Thank you. That's kind of what I thought.