harbourlab / SparK

Publication quality NGS track plotting
102 stars 20 forks source link

Represent the raw data for sparse points #20

Open ifantasy opened 2 years ago

ifantasy commented 2 years ago

This function of Spark is meaning for representing the type of data, especially for regions with intensive or clustered points. However, this function can not show real distribution for our data. We have another type of methylation. It occurs sparsely for some region, so it is different from WGBS. See the example below:

image

The raw bedgraph data related to the eight points is listed as follows: chr4 55527803 55527804 100 chr4 55527872 55527873 97.6374 chr4 55528238 55528239 88.1444 chr4 55530128 55530129 85.4848 chr4 55530492 55530493 93.8322 chr4 55530552 55530553 77.9618 chr4 55530798 55530799 88.2967 chr4 55532045 55532046 90.9048

You can see that the 1st and 2nd point have significantly distinct height, but 2nd and 7th points have similar height. Also think about the 8th point. So, I wonder if there is a way just to show the raw data?

My command is python $(which SparK.py) -pr chr4:55527304-55532546 -cf test.bedgraph -ps all -gtf gencode.vM25.annotation.gtf -o test

Originally posted by @ifantasy in https://github.com/harbourlab/SparK/issues/15#issuecomment-1023585157