Download MethylMix summary

yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts
no text concepts found
MethylMix summary
MethylMix is an algorithm that identifies functional differentially methylated genes
in cancer. It integrates DNA methylation from normal and cancer samples and
matched cancer gene expression data via a three-step algorithm:
 Step (i): Identifying the methylation state of each CpG site or cluster. We
applied univariate Gaussian mixture modeling to identify subgroups of
patients with similar DNA methylation level for a specific CpG site or cluster.
(A CpG cluster is defined as a set of highly correlated CpG sites for a single
gene). Each mixture component is referred to as a “methylation state” and
represented by its mean methylation level.
 Step (ii): Defining hyper- and hypo-methylated cancer CpG sites or clusters
relative to normal. To determine if a specific CpG site or cluster is hypo- or
hyper-methylated in cancer, we compare the methylation levels of each
methylation state (from step (i)) to the mean of the DNA methylation levels
of normal tissue samples, using a rank sum test.
 Step (iii) Identifying functionally relevant methylation. In MethylMix, each CpG
site/cluster is associated with its closest gene. MethylMix requires that the
DNA methylation levels of a CpG site/cluster has a significant effect on its
corresponding gene expression in order for the gene to considered a
methylation-driven gene.