I'm new to these type of analyses, so I wanted to check and see if someone could explain a couple things for me. I'm in the stages of PoolSeq using Nanopore to call the methylome to look for divergent regions of methylation.
I know that each means - average read depth, read quality, and assembly, but not how they fit together. The consensus is higher avg read depth -> genomic coverage -> greater experimental power. Quality being an assessment of how many reads we're successfully mapping, which we want to have as many '9's' as possible. And then there's also the ability to draft, polish, and compare assembly to reference to determine how well what we've sequenced matches up to what we currently know about the genome.
When it comes to methylation analysis, which is called based from the raw signal level data, how do these elements come in to play? Does assembly and assembly polishing have any impact here? What happens if the assembly quality is better than read quality? Are we just focusing on average read depth and read quality? What are some things we could be doing to raise our quality as high as possible from a methylation-aware standpoint?
Just one thing I'd point out is the difference between sequencing quality and mapping quality. Both have scores associated with them, but they mean different things. Sequencing quality is based on the confidence of correctly calling a given nucleotide. Mapping quality refers to the confidence of the read aligning to the reference.
In that case is there any benefit or drawback if someone was in a situation where their sequencing quality was like 96% but their assembly quality was high?
It's hard to say much without seeing the data, but it's probably fine. I recommend FastQC to check quality of the raw data.
Sorry that I can't help much with methylation analysis. I'm not familiar with that.