Custom Illumina primers

[taioba]

Hello everybody, we have just submitted a ddRAD genomic library for sequencing. We are interested in population genetics and this library has DNA from 186 individuals belonging to 6 different species. Our goal is to discover SNPs and usem them in popgen.

We used a combinatorial barcoding approach based o Adapterama I that employs one degenerate barcode on the i5 adapter to control for PCR-induced biases into the popgen stats, followed by two inline (R1 and R2) Illumina provided barcodes (indices D501-D508 and D701-D712, respectively, totaling 96 pairwise combinations) and two different Illumina barcodes on the i7 adapter. The protocol thus allows for a maximum of 192 barcode combinations, thus exceeding our sample size (n = 168).

We conducted size selection using magnetic beads and our library has an average size of 580bp.

Our goal was to get it paired end sequenced using a single NovaSeq 6000 S4 lane (300 cycles, 2x150bp).

The technician at the sequencing facility, upon reviewing our project, detected a single basepair missing from our R2 adapter.

There should be an extra A/T basepair at the very end of the TruSeq R2 sequencing primer annealing site.

Since this is the reverse primer, this would be right before our inline R2 barcode when the complementary strand of the adapter is read in the 5'->3' direction:

TruSeq2 Annealing Site (Adapter)

XXXXXXCTAGCCTTCTCGTGTGCAGACTTGAGGTCAGTG
<-3'-TCTAGCCTTCTCGTGTGCAGACTTGAGGTCAGTG-5'

TruSeq2 Sequencing Primer


This seems like a huge problem because we need all our barcodes in order to fully demultiplex our samples.

I was wondering if it would be possible to design a TruSeq R2 primer tailored to our adapter and use that in our run.

Alternatively, we could go for single end, sequencing but I guess that, given the average size of our library, most of our R2 barcodes would not be sequenced, greatly reducing the number of usable SNPs in our final dataset.

Any help is welcome!

Many thanks in advance!

Regards,

Marcos

[Ben3]

taioba I don't have experience with custom primers, but one thing I wondered is if that sequencing facility had a MiSeq or NextSeq 1000/2000. They are capable of longer reads and I wondered if you could do a single end 300 cycle run. It's not perfect, but I was thinking it could be a temporary fix.

[taioba]

Ben3, thank you so much for the tip. Given the fragment size distributions we got from BioAnalyzer, it looks like a significant fraction of our library would fall in the 250-300bp range. The facility has only a NovaSeq 6000 and a MiSeq. Could I go with the kit we were budgeted for (NovaSeq 6000 v1.5 300 cycles) and ask them to conduct a single end, 300 cycle run? Or is this sort of run only doable on a MiSeq or NovaSeq 1000/2000? My other option would be the (cheaper) 6000 SP v1.5 (500 cycles). Its output is not as high, but if we could sequence 500bp in one direction it would me covering most of our samples.

[SeqBuddy]

If you go the custom primer route I think NovaSeq supports that (see linked), but I do not think you can have a single S4 lane called out for it, the whole flow cell would need to be the same set-up. Maybe custom primers and SP or S2 flow cell would be an option?

- NovaSeq Series Custom Primers Guide (1000000022266) (illumina.com)
- Indexed Sequencing Overview for Illumina Systems