Hi, I'm new here. I got problems when I want to integrate the chipseq with rnaseq data. What I’m trying to do is quantificating chip signal of differential gene and presenting the results with a boxplot like the following picture.
Any help will be appreciated. Thanks advance!
You would first divide the RNA-seq data into groups based on your logFC criterium. Then get the coordinates of the TSS of these genes. Since genes have multiple transcripts you have to decide whether to use one or many TSS, or maybe the one with highest expression. That will give you coordinate(s) per gene. Next, use the ChIP-seq data and assign peaks to each gene, by whatever criterium you feel good with. Can be distance (every ChIP-seq peak within 10kb or so) to a gene, or anything else, it's difficult because there is little ground truth data. See the literature on peak-gene assignment methods. They all have caveats. In any case, once done you have per gene a list of ChIP-seq coordiantes. Then get the logFC of these coordinates. The x-axis is then the genes, the y-axis is the logFCs of the ChIP-seq datasets. In R the GenomicRanges package may help, in bash maybe bedtools intersect and bedtools closest. If you try something and then say where you get stuck one can try to help out with debugging.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
What data do you currently have? Do you have the log2FC values and your ChIP-seq values? If yes, you could manually bin your data based on your l2FC thresholds and make the plot that way.
Thank you so much for your help. I do have the log2FC and ChIP-seq values. But I have no idea how to do this. Maybe deeptools computeMatrix? Could you send me the link or tell me which command could bin mydata?