Hello,
I'm a neonate in NGS just starting to get into it. I'm trying to plan for a whole genome bisulfite sequencing experiment to determine methylation in a coral with a genome of 0.5 GB. My school recently purchased a MiSeq sequencer but every document I have found says that it is not enough to do what I'm planning. I want to know if this is accurate before I go and find an external source that will be more expensive.
Thanks,
Javier
Do you have a reference genome?
It might be difficult getting enough coverage given the output of the MiSeq, a HiSeq would be more suitable.
Another alternative would be to use some sort of reduced representation technique (RRBS, BsRADseq, etc.). This should help somewhat with the coverage but you'd be interrogating a smaller proportion of the genome. It has been shown to work quite well though. I'm not aware of anyone doing in corals so it might need some optimization in terms of restriction enzymes combinations and such.
A third option: Use a long read sequencer (i.e. Oxford Nanopore MinIon). You can actually extract methylation signals from the read traces without the need of bisulfite treatment. The output and error rate has improved a lot recently too. They're quite affordable and you could probably get enough coverage with a few flowcells. I have one in the lab if you want to come over and check it out. Let's talk!
Cheers,
Jorge
Javier, you really ought to be able to calculate this for yourself. A quick Google search shows that maximum output from a MiSeq is 25 million reads @ 300 bp per read. That's 7.5 Gb data, or 15 x coverage of your genome. This should be sufficient to give you an initial insight into the methylation status of most areas of your genome of interest. Nanopore Minion will definitely NOT give you the coverage you need unless you run many tens of flow cells.
The newer MinIon flowcells have a theoretical output of 10-20 Gb (it's still variable so realistically more like 5 GB), which is not that different from the MiSeq example while probably being more affordable. A MiSeq would be advantageous over ONT for other reasons (mainly quality), but output isn't one of them anymore. Once you factor in other costs (bisulfite treatment, library prep, replicates, etc.) The MiSeq really loses its appeal. If anything you'd be much better off in that sense using a HiSeq, which is hard to beat for cost-effectiveness.
- Badges
- Science method