Hi all,
I'm working on a set of GRO-seq data to calculate gene transcription rate. DRB was used during the experiment to release Pol2 and made a part of the original read band covering the genes at 0 min have some "blank". Then the transcription rate would be calculated by dividing the "blank" length by corresponding time period.
However, my question is how to define the boundary of the "blank". I have found a Bioconductor package groHMM, which claims be able to do that, but I noted that it originally performed on a set of hormone started transcription data, which has no read band at 0 min and the band length will increase as the time become longer. This situation is in contrast to my case of DRB involved experiment and this package may not suitable for me.
Hence, could anyone give me some suggestions? I know most paper analyze this by HMM model and the method seems very complex. Is there any packages or software?
Thank you so much.
Using an HMM for this seems like an overkill. The simpler idea would be to (1) calculate background coverage at time 0, (2) blacklist regions significantly above that at time 0, and then (3) find the extent in subsequent time points of regions significantly above the aforementioned background (ignore blacklisted regions). Use a large sliding window for step 3.