Miseq targeted resequencing - custom amplicons

[BDIOGG]

Afternoon all,

I'm attempting to design primers to enable me to run multiplexed custom amplicons on the Miseq, my lab is rather new to this and I have a few questions about the primer design.

I am using a protocol derived from the rather famous 16s sequencing protocol. So the primers include the illumina adaptors, gene specific sequences and the index sequence (on the R primer). Performed in a single PCR (i.e. not nested).

However, unlike the 16S protocol I am attempting to look for rare genetic variants in cancer cells. A concern I have is that I have no way of identifying variants that are just PCR error, if they occurred during an early PCR cycle there would be many copies present by the end and would be indistinguishable from true genetic variants (expected frequency is rather low). Unlike normal library prep methods, all of my sequenced reads will start from exactly the same point, so can't be confirmed in other reads with different start/end positions.

I thought perhaps performing the PCRs in triplicate, using indexed F primers would mean that if the rare genetic variants were identified in seperate PCRs then they are likely to be genuine. Sound sensible? If so, how would I go about indexing them? Can it be incorporated using a second index sequencing primer, similar in design to the first? Or could I add a 2/3 bp index to the end of the F primer, and sequence this using the Rd1 primer? Would this be a royal pain to decipher when it came to the bioinformatics (I have a bioinformatician on board, just need to know if I will be complicating his life significantly).

My library will end up very small (5 amplicons, 6 samples) and I want enormous coverage (i.e. 100k x if possible) so I will soon have questions about library complexity, coverage, etc etc however I think we have a better understanding of how to tackle that.

Thanks in advance.

[nucacidhunter]

With 5 amplicon x 6 samples x 3 replicate for each you get 90 amplicon species. The easiest way would be to use Illumina’s 16S protocol (http://supportres.illumina.com/docum...15044223-b.pdf).
You can use different index combinations from Illumina Nextera XT kit if doing separate PCRs or just use 18 different XT indexed primers for each separate sample and PCR replicates. If you sequence equimolar of amplicons with MiSeq V3 with 10% Phix with 75% cluster density of diverse libraries, it would yield around 15-20 M reads. With 15 M reads an estimated coverage of 160,000x/amplicon are achievable.