Hi,
I did peak calling analysis using MACS2, and I uploaded the results into UCSC genome browser. I get really weird looking graphs, they look like squares, not the regular hill shape. At first I thought it was diplicates in my data, so I used two methods to remove duplicates (Picard-tools and FASTX) but got these in both... any ideas? Thanks,
Try to understand what is given in this link
Just to prevent from redundancy you can read here as well.
Peaks are not tracks. Peaks show particular enrichment of your TF/histone marks against an input or for that matter no input , indicating that you have an enrichment. Having said that, you should generate the entire tracks of your aligned bam file for each of your samples and upload them in UCSC or IGV. This gives you the hill-shape profiles of the alignment files for both your ChIP for specific TF/histone and then below the second track for your input or control. Then you should upload the bed file of the peak which is ideally the blocks that span near a hill track peak. This shows that this region which has a signal boost also has the peak that corresponds to a rectangle block below the hill.
Your tracks represent your chip-seq profile for both TF/histone marks and your input. Having a peak under a hill shaped track is a testament that there was enrichment when you try to find signal enrichment of your particular TF against its input. I guess this should be clear for you. This is just a basic idea am giving you. There is more complex way to present output of ChIP-Seq as well but as of now try to understand what you see and why you see such tracks and what they stand for.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
I generated both bam and BW files, as well as peaks with bed files. the peak shape is just very linear/boxed in some places (not in all, but in quite a few). This is not a matter of technical difficulty with ucsc, but of some problem with the data I have or with the way I processed it.
Please post the image here so that we can take a look at it and help you farther.