Velvet Assembler: expected coverage versus estimated coverage versus effective covera

DMCH · 11-29-2011, 05:49 AM

Hello everyone,

I am very new to bioinformatics in general, and I am trying to assemble some short reads using velvet. Having read the manual, I am still not clear on the distinction between some of the coverage terms.

Specifically, is the expected coverage the length weighted node or contig or kmer coverage?

Is the estimated coverage actually referring to an estimate of the expected coverage, performed by velvet?

Is the effective coverage the node coverage after errors have been removed using the coverage cutoff? Or is it an estimate of the proportion of the genome length that is covered by the assembled contigs?

Finally, is there a difference between the coverage depth, and the coverage density? Are both these values expressed in units of 'X'? e.g. 5X coverage.

Many thanks indeed for any help. This is really frustrating me!

SES · 11-30-2011, 05:21 AM

Velvet coverage is not x-based. The coverage you are getting is k-mer coverage per contig. I know this question has been asked before, so I won't go into any detail. Some quick google searches should get you the answers you are looking for.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
de novo assembly using Trinity versus Velvet-Oases	Nol	De novo discovery	8	10-26-2013 12:56 PM
Get coverage of each site on contig generated by Velvet	genelab	Bioinformatics	5	08-03-2013 11:40 PM
Coverage (Velvet)	yasutake	Bioinformatics	6	08-07-2011 11:32 PM
Transcriptome assembler versus genome assemblers	samanta	General	9	08-02-2011 10:21 AM
Velvet and coverage range	BoggyP	De novo discovery	1	05-31-2011 10:46 AM

11-29-2011, 05:49 AM	#1
DMCH Junior Member Location: Oxford Join Date: Aug 2011 Posts: 2	Velvet Assembler: expected coverage versus estimated coverage versus effective covera Hello everyone, I am very new to bioinformatics in general, and I am trying to assemble some short reads using velvet. Having read the manual, I am still not clear on the distinction between some of the coverage terms. Specifically, is the expected coverage the length weighted node or contig or kmer coverage? Is the estimated coverage actually referring to an estimate of the expected coverage, performed by velvet? Is the effective coverage the node coverage after errors have been removed using the coverage cutoff? Or is it an estimate of the proportion of the genome length that is covered by the assembled contigs? Finally, is there a difference between the coverage depth, and the coverage density? Are both these values expressed in units of 'X'? e.g. 5X coverage. Many thanks indeed for any help. This is really frustrating me!

11-30-2011, 05:21 AM	#2
SES Senior Member Location: Vancouver, BC Join Date: Mar 2010 Posts: 275	Velvet coverage is not x-based. The coverage you are getting is k-mer coverage per contig. I know this question has been asked before, so I won't go into any detail. Some quick google searches should get you the answers you are looking for.