Different versions of samtools-pileup

dg.pooja · 02-11-2011, 07:36 AM

Hello,

** I have used pileup function with -vcf options both with samtools versions 0.1.7a and 0.1.12a

** I found the pileup output from both versions of samtools to be very different.

** Using the same aligner with exact same options, samtools 0.1.7a is giving me 452 unique SNVs and the version 0.1.12a 251 unique SNVs.

** There are 196 SNVs that are common to both versions of samtools. This is after I use varfilter with default parameters with SNP quality cutoff of 20.

** I understand that the old version of samtools use the MAQ SNP calling criteria and we have not used -A option with samtools version 0.1.12a, hence have not forced MAQ calling criteria.

*** I am confused with the data and would like to understand why there is such difference in results between the two versions of samtools.

Can any one help in this regard ?

Thank you
Pooja

drio · 02-11-2011, 08:19 AM

The snp caller engine has changed dramatically among those versions. Review carefully the manuals, particularly mpileup.

That being said. The overlapping events sounds a little bit low but it depends on the coverage for your project.

swbarnes2 · 02-11-2011, 09:39 AM

Using the -B option on the new pileup command might cause it to behave more like the old pileup.

drio · 02-11-2011, 09:46 AM

If I am not mistaken the latest version uses the SOAP model instead of the MAQ model at the core of the snp caller. Let us know if by removing BAQ computations (-B) you get similar results to the old version.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
samtools pileup	BioSlayer	Bioinformatics	9	01-16-2012 04:10 AM
samtools pileup	raghu1990	Bioinformatics	0	06-28-2011 03:26 AM
samtools pileup	wisosonic	Bioinformatics	0	05-11-2011 05:57 AM
Tophat v1.1.4 bug - cannot handle non-numeric versions of Samtools	adumitri	Bioinformatics	4	01-10-2011 06:18 AM
samtools pileup format	foxyg	Bioinformatics	3	11-16-2010 08:25 AM

02-11-2011, 07:36 AM	#1
dg.pooja Junior Member Location: India Join Date: Nov 2010 Posts: 2	Different versions of samtools-pileup Hello, I have used pileup function with -vcf options both with samtools versions 0.1.7a and 0.1.12a I found the pileup output from both versions of samtools to be very different. Using the same aligner with exact same options, samtools 0.1.7a is giving me 452 unique SNVs and the version 0.1.12a 251 unique SNVs. There are 196 SNVs that are common to both versions of samtools. This is after I use varfilter with default parameters with SNP quality cutoff of 20. I understand that the old version of samtools use the MAQ SNP calling criteria and we have not used -A option with samtools version 0.1.12a, hence have not forced MAQ calling criteria. * I am confused with the data and would like to understand why there is such difference in results between the two versions of samtools. Can any one help in this regard ? Thank you Pooja

02-11-2011, 08:19 AM	#2
drio Senior Member Location: 41°17'49"N / 2°4'42"E Join Date: Oct 2008 Posts: 323	The snp caller engine has changed dramatically among those versions. Review carefully the manuals, particularly mpileup. That being said. The overlapping events sounds a little bit low but it depends on the coverage for your project. __________________ -drd

02-11-2011, 09:46 AM	#4
drio Senior Member Location: 41°17'49"N / 2°4'42"E Join Date: Oct 2008 Posts: 323	If I am not mistaken the latest version uses the SOAP model instead of the MAQ model at the core of the snp caller. Let us know if by removing BAQ computations (-B) you get similar results to the old version. __________________ -drd

02-11-2011, 09:39 AM	#3
swbarnes2 Senior Member Location: San Diego Join Date: May 2008 Posts: 912	Using the -B option on the new pileup command might cause it to behave more like the old pileup.