What is the best approach to perform ultra-deep targeted sequencing of few samples?

[jocognito]

Any help would be appreciated?

Our group have performed recent whole genome/exome sequencing in tumor samples and now I would like to take a few of these variants (say maximum 50 amplicons) to:

1) Validate them in the original samples
2) Perform ultra-deep sequencing in an extension cohort of samples but using different amplicons for different samples (hope that makes sense!)

How would one do this...as majority of the targeted sequencing approaches focuses on a large panel of genes/exons in a large number of samples. We have experience of using Fluidigm but I don't need to multiplex that number of amplicons.

So for example, say I want to screen just 2 exons of KRAS in 10 samples but want a coverage of say 10,000x so I can study the clonal complexity and evolution. How would one do this?

Thank you!

[SNPsaurus]

There are a few ways to do that. One would be to order pulldown oligos for the exons of interest (for example, IDTI lockdown oligos https://www.idtdna.com/pages/product...ockdown-probes). The workflow would then to fragment your genomes and barcode (nextera or other kit), then pulldown the exons from the samples in a single reaction, then amplify and sequence. You are likely to have very high off-target reads, since you are aiming for such a small portion of the genome, but putting those samples even on a MiSeq would likely get 10,000sX coverage.

Alternatively, you can amplify each exon by PCR. Then pool amplicons by sample, fragment (Nextera XT) and barcode, amplify and pool for sequencing. You'll get higher coverage but it is more laborious. So depending on the # of exons and samples one would be better than the other.

My academic lab does paired-end sequencing of short DNA fragments in order to detect variants in the 1 in 5000 range and are working on some tumor clonal samples right now. Feel free to contact me at [email protected] if that is of interest.

[mchizar]

What aligner are you will you be using? Aligning and gapping sequence data that is 10,000X deep can be a challenge. DNASTAR's SeqMan NGen assembler can align data this deep and produces a BAM file output that can be visualized and analyzed for variations. Not sure if this is possible with a genome browser/viewer when data is this deep.