WGCNA analysis for bisulfite data.

Question

[archi]

I want to use Weighted co-relation network analysis (WGCNA) for bisulfite data (DNA methylation study) . I got seven treatment with three replicates each. (7*3= 21 samples). I have done WGCNA for RNASeq data and I understand we need normalised matrix count as an input file. What would be the input file for bisulfite data? Has anyone done similar study, could you please help me?

Many Thanks Archana

[Chip]

I haven't done this kind of analysis myself, but bisulphite sequencing data consist of reads. You must align the reads and analyze them with tools such as bismark. At the end of the process you will obtain a matrix of Beta-scores (representing percentage of methylation at different CpG sites).

Answer 1

[archi]

Thanks Nick!

I have used bismark for alignment and methylation calling. This gives me coverage per site but not the beta value.

[Chip]

Did you use bismark_methylation_extractor? I see from the user guide that the --bedGraph flag will output what you need.

I inaccurately implied that bismark will create a matrix of beta-values but that's a step you have to perform binding data from different samples. (Sorry for being meticulous but you should reply to my comment instead of creating a new answer.)