Hi,
I am using QSEA package to analyse MBD-seq data on PC3 (prostate) cell lines. From the MBD-seq protocol I have samples generated from elutions at high salt concentration (2000mM) which aims at capturing highly methylated fragments and samples eluted at lower salt concentration (450mM) which is aiming to capture low methylated fragments (assuming low at CpG density). Foe each type of elution I have biological replicates.The problem is: When i run both types of samples in QSEA analysis using blind calibration, I seem to not get Beta values for samples at lower salt concentration (strangely this only happens to one biological replicate) and I get Beta scores for the unenriched sample (quite high scores actually)? Also, when I run the same samples using TCGA calibration I don't get Beta values at all in any of the samples generated from low salt concentration elution but again and I get Beta scores for the unenriched sample?
The only explanation i can think of is that those samples at low salt concentration may not be enriched?
Please would you be able to comment on this as I am worried that I may be doing something wrong!
Thank a lot, Ermira
Hi Ermira,
sorry, I somehow did not receive the notification of your post, and missed it. If it is still relevant, please provide more details. What is the unenriched sample? What exactly did you expect, and what is the result? For TCGA calibration, what exactly did you use? The normal samples of the PRAD cohort? There is also DNAme array of LNCaP clone FGC from ENCODE, which may be appropriate.
QSEA provides some QC metrics and plots to assess MBD enrichment, but the method depends on proper calibration. Have you looked at the plotEnrichmentProfile function?
Best, Matthias