DNA fragmentation and ION torrent

[ferjusa]

Hi all,

I'm new to the forum and also new to NGS.
I would like your opinion on the approach to be followed for an experiment with the ION torrent.
The idea is to sequence the genome of the HCV virus (approximately 9000 bp) or at least 3 genes of the virus (each of about 1500 bp).
How it’s better to fragment the DNA into smaller sequences? Which could be the method to follow? Avoiding the sonication, how can I proceed with an enzymatic fragmentation? And which is the better way to design the primers?
Could anyone give me some ideas? I really appreciate it!
Thanks so much!

[molbionerd]

I think sonication is probably the way to go with DNA as it is the least prone to bias. However, if you are set on using an enzymatic approach, you should be able to do it one of two ways. Either use DNase I and cut your genome for different time periods to determine the best timing to hit your target target size range. Another way would be to use restriction enzymes to digest. Has this virus been previously sequenced? If so then you should be able to determine the number of fragments generated by various restriction digests. This will inherently give you bias as you won't be able to sequence across any of the restriction sites, but if you follow up the digest with end repair you should be able to get half the digestion site and therefore be able to figure out what should have been on the other side. You can do targeted sequencing of it as well using primers designed to your virus (this assumes it has been sequenced previously). Agilent offers targeted resequencing library prep kits for relatively cheap and they should be able to help you design the primers that you need.

To be honest though, the fact that you have such a small genome should mean you could even do this using standard sanger sequencing from clones and around 20 to 25 clones would get you full coverage. If you need to be able to get greater depth then NGS will work. I would look into barcoding different samples/fractionation methods so that you can minimize the biases that will sneak in using any fractionation method.

I hope this helps.