laurahspencer / DuMOAR

0 stars 0 forks source link

Incorporate MiSeq data from MBDBS libraries in meth. analysis? #16

Closed laurahspencer closed 1 year ago

laurahspencer commented 1 year ago

I could concatenate the MiSeq data to the MBDBS-Seq data before aligning, etc., to incorporate more reads/data. Thoughts?

kubu4 commented 1 year ago

I think you might as well use the data. However, I'm still not convinced you should concatenate FastQs of mixed read lengths, different sequencers, etc.

What would be the rationale for concatenation?

mgavery commented 1 year ago

I don't know that it's worth it given the different read length/sequencer type. It's roughly 500k reads per sample - not much considering ~17 millions reads per sample from HiSeq