Hi all,
I've been working with whole genome bisulfite sequencing data from brain tissue. Our model uses a Cre-lox strategy to knock out Dnmt3a in a specific type of neuron, and we're now comparing genome-wide CpG methylation in control and floxed mice.
While we don't have any particular genes or genomic contexts we expect to see changes in (which is why we're doing WGBS), we expect at minimum for the knockouts to have less methylation. As it happens, the opposite is the case, and the knockouts pretty clearly have more CpG methylation. Genome-wide the average CpG methylation is a little less than 75% in controls, and around 76-77% in knockouts.
This is my first time doing such a study, so I'm wondering if anyone can offer some advice on technical reasons that I might be getting these results? What QC measures could give me a hint that the library prep went wrong, or that some kind of bias has entered the equation. It's possible that these results are real, but since they're weird I want to be extra sure no technical errors are the cause.
Thanks!
I've been working with whole genome bisulfite sequencing data from brain tissue. Our model uses a Cre-lox strategy to knock out Dnmt3a in a specific type of neuron, and we're now comparing genome-wide CpG methylation in control and floxed mice.
While we don't have any particular genes or genomic contexts we expect to see changes in (which is why we're doing WGBS), we expect at minimum for the knockouts to have less methylation. As it happens, the opposite is the case, and the knockouts pretty clearly have more CpG methylation. Genome-wide the average CpG methylation is a little less than 75% in controls, and around 76-77% in knockouts.
This is my first time doing such a study, so I'm wondering if anyone can offer some advice on technical reasons that I might be getting these results? What QC measures could give me a hint that the library prep went wrong, or that some kind of bias has entered the equation. It's possible that these results are real, but since they're weird I want to be extra sure no technical errors are the cause.
Thanks!