Questions about RADseq oligo

[barn88]

I will be genotpying ~500 samples using the original RADseq protocol, and I have a couple of questions.

First, is desalting a sufficient purification method for the barcoded adapters? I plan to order P1 adapters with 48 different barcodes, and other purification methods add significantly to the cost so I'm wondering if the added cost is worth it.

Second, I would like to use combinatorial multiplexing by using barcodes for individual samples plus an index to identify separate pools of samples. In ddRAD, the index is incorporated into one of the PCR primers, but in the original RAD protocol the PCR primer is shorter and has no appropriate place for an index.

Does anyone have any suggestions for how to modify the original RADseq protocol to allow combinatorial multiplexing?

[Carcharodon]

Originally Posted by barn88

I will be genotpying ~500 samples using the original RADseq protocol, and I have a couple of questions.

First, is desalting a sufficient purification method for the barcoded adapters? I plan to order P1 adapters with 48 different barcodes, and other purification methods add significantly to the cost so I'm wondering if the added cost is worth it.

Second, I would like to use combinatorial multiplexing by using barcodes for individual samples plus an index to identify separate pools of samples. In ddRAD, the index is incorporated into one of the PCR primers, but in the original RAD protocol the PCR primer is shorter and has no appropriate place for an index.

Does anyone have any suggestions for how to modify the original RADseq protocol to allow combinatorial multiplexing?

1) I've done ddRAD, and a couple of other folks I know have also done ddRAD, and we just use standard desalting. The HPLC purification that is often recommended is considerably more expensive, and we've had good results without it.

Other methods of purification, like HPLC, make sure that nucleotides on the 5' end of an oligo aren't cut off during synthesis (rather, it makes sure that those that WERE cut short on the 5' end don't make it into the final product). That might be a consideration for T-overhangs that are used for the P2 adapter for RADseq (and might be less of a consideration for the longer overhangs in ddRAD). But others can chime in on that.

2) Why not just design your P2 adapter and primer after the Peterson et al. (2012) protocol? Literally the only thing you'd have to do differently is take their P2 adapter and change the sticky-end to append to an A-tail. Or am I missing some nuance here?

[barn88]

Thanks for your response. I had considered modifying the ddRAD P2 adapter; I just wanted to see what suggestions other people made.