Sequencing microsatellite enriched DNA amplicons on MiSeq

[Polar314]

Hello all,

I am trying to figure out how to sequence amplicons I have made using a microsatellite DNA enrichement protocol.
Briefly, I took genomic DNA from an animal, digested it with restriction enzymes, ligated adapters on the fragments, hybridized them with biotinylated oligos with repetitive elements, recovered the hybridized DNA with streptavidin beads, and recovered the enriched DNA with PCR.
So now I have multiple amplicons from each animal, with an average length of 500bp, and each amplicon has a known adapter sequence on each end.
Illumina doesn't have a kit that supports this protocol, so I am trying to figure out what will be the best way to load my samples on a MiSeq for PE reads. I will also need multiplex several (6) samples on one flowcell.
Will the TruSeq sample prep kit work, if I skip the end repair stage (as my amplicons are blunt ended) and proceed from the adenylation stage?
Any input will help! This NGS is all new to me.

[matth431]

Hi,

The issue I suspect you will have is that the presence of the first adaptor sequence from your enrichment will confuse the cluster identification step of TruSeq chemistry... MiSeq and HiSeq both look at the first 4bp of sequence to determine whether close "spots" of DNA on the flow cell are separate molecules. If they don't differ across these nucleotides, they will be assumed to be one sequence... if they actually aren't, you'll essentially get a double-read effect and they will later fail QC and be removed. Since from the sound of it, ALL of your amplicons will begin with the same 4bp, you should expect the majority of clusters to be thrown out. I've read some suggestions that spiking in 50% PhiX control can help with this, but then you'll need to account for only getting half as much useful data as you wanted.

One option to avoid this in future would be to convert the custom adaptors for genotyping-by-sequencing (i.e. RAD sequencing) to act as your enrichment adaptors (add a biotin tag to one of them). This would mean you could have barcode sequences as your first 4bp of sequence... still low-complexity compared to genomic DNA, but if you pool enough barcodes together it would be more diverse than having a single adaptor sequence as you have here.

Using the TruSeq kit on the amplicons should work, but you'll lose some informative sequence at each end because you'll need to sequence through your adaptor... but if your amplicons are ~300bp anyway you should still cover them with 2x150bp reads.