View Single Post
Old 04-25-2018, 03:13 PM   #8
SNPsaurus
Registered Vendor
 
Location: Eugene, OR

Join Date: May 2013
Posts: 451
Default

We got good results with both NextSeq and HiSeq4000 runs, so I think you are safe either way. Given the size of the genome, I'd want as much sequence information at each locus to help improve the mapping accuracy. I can't quite remember how the NextSeq does these days with invariant nucleotides (the cut site). It used to be an issue and require lots of phiX. I think they fixed it to some extant but I'd be careful if running the lane yourself and look into it.

Will you pair the methylation-sensitive and insensitive libraries for every sample? Given ddRAD's propensity for locus drop out (from size selection variation and SNPs in the cut sites and locus sequence) you will need to have a control library for each sample and it should be in the same size selection run.

Can you share details on how you got the 100k sites from the simulation? That one still has me wondering.
__________________
Providing nextRAD genotyping and PacBio sequencing services. http://snpsaurus.com
SNPsaurus is offline   Reply With Quote