title	author	date	output
Methylation Site Identification	Bharath Manicka Vasagam, CCHMC, The Barski Laboratory	March 8, 2016	html_document

##Objective The objective of this code is to to comapre the experimental methylation site output to that of a control of Mus musculus and identify the number of

1. Number of CPG islands in the experimental data.
2. Average of the % methylation in the experimental data.

##Packages Used

1. data.table
2. GenomicRanges
3. rtracklayer
4. plyr

Introduction

The methylation site identification code helps to compare the experimental promoter methylation sites output with that of the control (with all methylation site details) of Mus musculus.

Conversion of files to BED files using LIFTOVER:

The experimental output file was in TXT format. It was converted to a standard BED format as ChrN:Promoter Start - Promoter End using R.

Then this csv file was uploaded to liftover https://genome.ucsc.edu/cgi-bin/hgLiftOver

The BED output file from liftover was then data cleaned, compared with the control to find the related overlaps from the GenomicRanges package.

groupA<- data.table(exp_data)

#setting the table key
setkey(groupA, chr, promoter_start, promoter_end)
groupB<- data.table(bgwg)

#setting the table key
setkey(groupB, chr, start, end)

#overlapped files
over <- foverlaps(groupA, groupB, nomatch = 0)

Result:

Thus after identifying overlaps, number of CPG islands and % average methylation in the experimental data was calculated. To ensure data integrity, manual calculation of CPG islands was performed for few sanple promoter regions.