SEQanswers (
-   Bioinformatics (
-   -   Get coverage value in given regions (

AndreaIzq 07-03-2013 04:08 AM

Get coverage value in given regions
I am trying to get the normalized coverage value in some given regions (bed file) but I don't succeded. Is there any easy way to do so? I am quite new in bioinformatics, sorry for asking basic questions.

Thanks in advance,


Heisman 07-03-2013 07:22 AM

What do you mean by normalized? Do you want the raw coverage counts for different samples and then to just divide by the total number of reads? Do you want coverage of every base in a region or just the average coverage within a region? What commands have you tried to achieve this?

AndreaIzq 07-03-2013 08:20 AM

Thanks Heisman for your reply.
What I exactly want is to plot the average coverage of my proteins in several regions (bed file). In the y axis, the average coverage and in the x axis, the width of this regions.
I want to know if there is a direct relationship between the width of that regions and the coverage of my proteins. I don't know what normalized procedure to use to be able to compare the coverage of my proteins.

Thanks again,

Heisman 07-03-2013 12:53 PM

Is this an RNA-seq experiment? Have you heard the term "RPKM", and do you feel that would be appropriate regarding normalization?

I haven't worked with RNA-seq data myself so maybe someone else will chime in with a tool that can do what you want without issues.

Without incorporating any normalizing, though, if you just wanted to compute average coverage for a series of regions, I believe the easiest way to do it is described here:!

Which shows a variant of this command:

coverageBed -abam ALIGNED_FILE.bam -b REGIONS.bed -d | groupBy -c 5 -o mean

AndreaIzq 07-04-2013 02:57 AM

That's a ChIP-Seq experiment. I wil try to do it with the average coverage as you suggested, if the results are not correct I will try RPKM then.


Heisman 07-04-2013 08:32 AM

I have not done ChIP-seq so I may be missing something but the only thing that jumps out as obvious for normalization purposes is the number mapped reads for each sample. Good luck.

All times are GMT -8. The time now is 09:42 AM.

Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2019, vBulletin Solutions, Inc.