calculating the coverage

lyz1030 · 04-03-2012, 09:22 AM

Hi all

If there are two reads convering a position, and one read happens to have a delete in this position. I want to know the coverage (or the read depth) in position 5, 1 or 2? What is the difference between the coverage and the read depth? If the quality is below the base quality threshold, what is the coverage then?

123456789
AGGGTCTCC
AGGG.CTCC

Thanks.

twaddlac · 04-03-2012, 09:47 AM

There are a number of tools that can do this, but the two software suites that are widely used are samtools (http://samtools.sourceforge.net/) and GATK (http://www.broadinstitute.org/gsa/wi...alysis_Toolkit) and picard (http://picard.sourceforge.net/comman...overview.shtml)

Unfortunately people use depth and coverage, indicating that there is 10x coverage of nth base, alternatively 10x depth of nth base. The way that I use them is depth is used to describe how many reads are covering nth base and coverage longitudinal in, e.g. 90/100 bases of the reference are covered.

Depth of coverage is also a term that is used to describe the average depth of each base across the reference.

The result when calculating the coverage at a low quality base is dependent upon the software and its settings. If you lower the cutoff then you'll see it. They also have a threshold for mapping quality which may be of interest to you.

lyz1030 · 04-03-2012, 12:03 PM

Thanks. I want to know if the 'delete base' should be excluded when calculating?

twaddlac · 04-03-2012, 06:03 PM

Well that depends on your study. If you're looking for a variant then I would keep in there because, depending upon the base and map qualities, it may be a true variant. However, if you're generating mapping statistics (depth, coverage, depth of coverage) then I would not.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Newbie Question: Calculating physical coverage from genome coverage	tristanstoeber	Illumina/Solexa	4	06-24-2013 11:53 AM
calculating mismatch rate	adrian	Bioinformatics	0	03-15-2012 07:05 AM
what program can i use for calculating coverage?	dkrtndhkd	Bioinformatics	2	02-03-2012 06:39 PM
Gene list for calculating coverage	nseh	Bioinformatics	1	05-22-2011 07:31 AM
calculating expected reads	Seqasaurus	454 Pyrosequencing	2	09-08-2010 04:27 AM

04-03-2012, 09:22 AM	#1
lyz1030 Member Location: USA Join Date: Mar 2011 Posts: 33	calculating the coverage Hi all If there are two reads convering a position, and one read happens to have a delete in this position. I want to know the coverage (or the read depth) in position 5, 1 or 2? What is the difference between the coverage and the read depth? If the quality is below the base quality threshold, what is the coverage then? 123456789 AGGGTCTCC AGGG.CTCC Thanks.

04-03-2012, 09:47 AM	#2
twaddlac Member Location: Pittsburgh, PA Join Date: Feb 2011 Posts: 49	There are a number of tools that can do this, but the two software suites that are widely used are samtools (http://samtools.sourceforge.net/) and GATK (http://www.broadinstitute.org/gsa/wi...alysis_Toolkit) and picard (http://picard.sourceforge.net/comman...overview.shtml) Unfortunately people use depth and coverage, indicating that there is 10x coverage of nth base, alternatively 10x depth of nth base. The way that I use them is depth is used to describe how many reads are covering nth base and coverage longitudinal in, e.g. 90/100 bases of the reference are covered. Depth of coverage is also a term that is used to describe the average depth of each base across the reference. The result when calculating the coverage at a low quality base is dependent upon the software and its settings. If you lower the cutoff then you'll see it. They also have a threshold for mapping quality which may be of interest to you.

04-03-2012, 12:03 PM	#3
lyz1030 Member Location: USA Join Date: Mar 2011 Posts: 33	Thanks. I want to know if the 'delete base' should be excluded when calculating?

04-03-2012, 06:03 PM	#4
twaddlac Member Location: Pittsburgh, PA Join Date: Feb 2011 Posts: 49	Well that depends on your study. If you're looking for a variant then I would keep in there because, depending upon the base and map qualities, it may be a true variant. However, if you're generating mapping statistics (depth, coverage, depth of coverage) then I would not.