Re: low DMC and qvalue

[Alexander Blume]

Hi Selma,

What exactly do you mean by "low DMCs"? Are you trying to find CpGs where the methylation difference is low (less than 20%)? 

You can not use getMethylDiff for this, since this will set a lower, not an upper boundary on the methylation difference. 
You could convert the methylDiff object into a data.frame using getData() to filter and sort it manually to find your sites with low methylation difference but high confidence, if any exist. 

I hope this helps.

Best,

Alex

leulmi selma schrieb am Montag, 17. Juni 2024 um 15:58:45 UTC+2:

Hi,

I am interested in low DMCs,  but when I filter the data to only keep CpGs with a methylation difference<20, I always end up with quite high qvalues for these CPGs. Any advice on how to analyse low DMCs with high confidence ? Many thanks.

Selma.