[BioC] assistance to understand MEDME

[mattia pelizzola]

Hi,

in the paper you can find that the MeDIP-chip enrichment is not
linearly related with the DNA methylation level.
MEDME is able to model this non-linear relationship using a
calibration dataset where the genomic DNA is fully methylated.
Limma is not used in MEDME. The only point where limma is mentioned in
the paper is for the array data normalization.
In this regard, enrichment data analyzed with MEDME should be already
normalized by the user.

I hope this helps,

mattia


---------- Forwarded message ----------
From: Ashwin Kumar <ashwin.havoc at gmail.com>
To: bioconductor <bioconductor at stat.math.ethz.ch>
Date: Thu, 23 Apr 2009 16:36:31 +0530
Subject: [BioC] assistance to understand MEDME
Hi Pelizzola,
                    You have mentioned in your paper (Pelizzola et al,
Genome.Res 2008) that DNA methylation data is not linear and you have used
limma in your data analysis. So i want to know have you fitted lmFit() on
your data? If so how MEDME will make change over linear modeling fitted data
? If i am wrong any where please excuse me.

Thankfully
A.Ashwin
Departement of Biotechnology
MLSC
Manipal University
INDIA