Hi helpful people,
I have paired-end sequence data for pooled samples which were library prepped with a custom inline barcoded P5 adapter. So the first 6 bases of the R1 read should contain the barcode for a sample. And it does, which is great. BUT for some samples up to 5% of the paired R2 reads also contain the barcode in the first 6 bases!
So for up to 5% of read pairs we have:
R1 read id 3001:
[barcode] [sequence............]
R2 read id 3001:
[barcode] [sequence............]
Of course we would expect by chance that some of the R2 reads would have their first 6 bases the same as the corresponding barcode. However not 5% !!
Any ideas what could cause the barcode to sometimes appear at the start of both R1 and R2 when only one barcoded adapter was used in library prep?
cheers
David
Hi Alex,
the quality of the R2 was only a bit lower across the read than the R1. In general the quality is fine, though the first 5 bases are clearly a bit lower in quality.
I am wondering if the barcode P5 adapters could have formed some sort of chimeric read with the P7 such that the R2 read begins with a barcode? However I can't visualise the fragment structure that would result in this happening...
since it is whole genome data, I would expect the barcode to appear randomly at the start of the read only a certain % of times. If a sample has, say, 11 million read paris, then by naive calculation we would expect (1/4^6) * 11m = 2,685 reads with that barcode at the beginning of the R2 read by chance. However I see about 500,000! So there is definitely something fishy.
I agree that it may be an artifact of resynthesis. Like you I am no illumina expert - anyone else out there can chime in?
- Badges
- Science method
- Similar topics
- Biological Science
- Genetics
- Genetic Analysis
- genetic techniques
- Sequencing