Tools for investigating the correlation of differential gene expression & histone marks/ TF binding?

Question

[chiefcat]

Hi everyone.

Would you please share some tools which can integrate differential gene expression & differential ChIP enrichment data?

I have both types of data from 2 time points (Stem cells vs differentiated cells) and I would like to investigate if the change of histone mark enrichment correlates with the differential gene expression in these two cell types.

Many thanks!!!

Answer 1

[shoujun.gu]

may be you can try bedtools intersect to get the overlapped region between RNASeq result (eg. -1000bp upstream) and ChIPSeq result.

[chiefcat]

Hi Shoujun. Thanks, I am looking into the tutorial on their website now. May need to ask you some more questions later! =] Thanks!

[chiefcat]

Hi Shoujun, I've just got the Cuffdiff output found the labmate who help me to handle the RNA-seq. data, what kind of corrdinate I should use for the bedtools intersect? Do you have any examples in your hand? Thanks!

[shoujun.gu]

The easy way is to use ChIPSeq annotation tools (like PAVIS) to get the gene list and then compare with your RNASeq DEG list. Or to expand the region of your RNASeq DEG list with bedtools slop, and intersect with your ChIPSeq peak.

for bedtools slop usage and detailed coordinate, it's really depends on your specific goal. You should check http://bedtools.readthedocs.io/en/latest/content/tools/slop.html

[chiefcat]

Thanks Shoujun, what kind of tools can be used for the visualization of the intersection result? Do you just usually present it using Venn diagram other any other better way which can really show the changes of all genes between two samples?

[shoujun.gu]

usually I just use very standard visualization methods, like heatmaps for DEG and igv for peak examples. Maybe there is some fancy tools, but I'm not familiar with that.

Answer 2

[amy.bashir]

I think BETA (http://cistrome.org/BETA/) is well suited for this task.

[chiefcat]

Hi Amy, thanks for your reply. I've read a little bit about BETA but the example on their website is too simple. In the test example, only one bed file is input for the analysis. In my case, should I input the differential enriched peaks between the two time point of samples? Thanks.