RAD-seq adapter design

[flobpf]

Hi

I was planning a RAD-seq experiment for about 90 ecotypes. The costs of ordering 180+ HPLC purified and modified oligos for those many samples are quite prohibitive. Hence, I was thinking of a double-index strategy for pooling all my samples using Illumina, and I wonder if someone could provide some feedback on whether the following oligo design would be appropriate for Illumina and RAD-seq.

So essentially, the P1 adapter would be

Read 1 sequencing primer -- Barcode -- REoverhang

ACACTCTTTCCCTACACGACGCTCTTCCGATCT -- AACTA -- CAT*G

The P2 adapter would be

Read 2 sequencing primer -- IlluminaIndex -- P7 attachment site

so that the final DNA would look like:

Read 1 sequencing primer -- Barcode -- REoverhang --DNA of interest --Read 2 sequencing primer -- IlluminaIndex -- P7 attachment site

I would also have to order two primer oligos for PCR amplification:

P5 attachment site -- Read1 SeqPrimer --> 1 primer
P7 attachment site -- IlluminaIndex -- Read2 SeqPrimer --> 10 primers

based on the information provided here

Thus, if I do Paired End sequencing, then for 90 ecotypes, I calculated that I would just need to order

P1: 10*2 = 20 (*2 to make double strand)
P2: 10*2 = 20
PCR primers: 11
Total=51

instead of ~200 odd oligos for the regular RAD-seq protocol, leading to significant cost savings. One drawback is that I won't be able to pool the samples after attaching the P1 adapter; instead I can only pool in sets of 10. However, given the cost savings, I think that is ok.

Does this strategy sound feasible?