A few tools for manipulating methylation call files

Script to merge multiple methylKit files together, e.g., from multiple replicates of one library.

Usage:

cat (or zcat) *.methylKit<.gz> | sort -k2,2 -k3,3n | merge_methylKit.py --prefix <prefix for output file>

OR

paste -d "\n" <(cat file1) <(cat file2) <(cat file3) <etc.> | sort -k2,2 -k3,3n | merge_methylKit.py --prefix <prefix for output file>

I'm not sure which is most efficient.

Will create <prefix>_merged.methylKit as the output.

Script to merge the information from both strands of a CpG dinucleotide. Comparable to MethylDackel extract's --mergeContext option

Usage:

cat <methylKit file> | merge_cpg_strands.py > <output file>

Script to downsample the observed methylation calls from a methylKit (or bedGraph) file.

Usage:

cat <methylkit file> | grep -v chrBase | downsample_methylKit.py --fraction <fraction of reads to keep> > <output file>

OR

cat <bedGraph file> | grep -v track | downsample_methylKit.py --fraction <fraction of reads to keep> --bedGraph > <output file>

This can be used in place of downsampling reads from a fastq/bam before doing methylation calling. The figures below show the similar results between downsampling before or after calling methylation:

Histogram of coverage per site:

Histogram of percent methylation per site:

Histogram of the difference in inferred methylation when downsampling before or after methylation calling: