I have a genelist with RNA-seq expression + logFC and etc. I also have Chip-seq peaks, how do I merge these two very different data sets?
Thanks!
I have a genelist with RNA-seq expression + logFC and etc. I also have Chip-seq peaks, how do I merge these two very different data sets?
Thanks!
You have to be a bit clear with your question. It depends on what kind of ChIP is performed? Is it for specific transcription factor or chormatin marks? It has to be species specific. You can try to see if your list of genes have transcription factors or not and that overlaps with the TF ChIP data which clearly shows that on doing ChIP-Seq on a particular TF for the same whose RNA-Seq was done you can find genes that are controlled. Or it can be Chromatin marks for which ChIP-Seq is done and different chromatin mark might act in either repressive or activation. In that case you can pull out the peaks that are under the specific chromatin marks, for promoters annotate them to specific promoter windows around TSS and extract list of genes or refseq ids and then overlap with your RNA-Seq gene list. This will enable you to understand to what extend the genes expression is being controlled by specific chromatin marks at the promoter level.
You can annotate the peak bed file for your corresponding species and extract the gene names or refseqID for pre define promoter windows in case you have refseqID as gene names for your RNA-Seq and then use any string venn diagram online tools like BioVenn or Venny to find the overlap.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
What do you mean by merge? Append the data into the same file? If you could give us a small example of your two data sets and a example of the final output someone here will be able to help you.
Merge peaks or find intersecting regions between the peaks of the Chip-seq data and also the genelist.
bedtools intersect is probably what you're looking for. You'll need to have the chr, start, and end information for each gene. And then do something like