R Methy VI: Analysis

Now you will compare the two groups to identify sites and regions which are differentially methylated in the patients and control samples.

site level analysis

First you need to extract the beta values from the normalized data, you also need to extract the phenotype information.

• Use the function getBeta() to extract beta values from minfi.norm
• Extract phenotype with status <- phenoData$Sample_Group
• Use the function dmpFinder( , , type= ) and fill in the blanks.

Now use the normalized data and perform a DMP analysis with status (case vs control) as phenotype.

Question

8. How many CpG sites have a q-value below 0.05?

Regional analysis

Perform a DMR analysis with status as phenotype, with a difference of 0.2 in beta values between the two groups.

• create a design matrix using design <- model.matrix(~status)
• Perform a DMR analysis using bumphunter( , design= , cutoff= )

Questions

9. How many DMR's did you find?
10. How many of the DMR's are on chromosome 6?

---

Next: ChAMP

Previous: Normalization