VCF file - no ALT but DP4 alternative exist

Rachelly · 04-03-2012, 05:23 AM

Hi all,

I'm using samtools to find variants in my resequencing data, and I've noticed something weird:

After running the following command:

Code:

samtools mpileup -L 10000 -d 10000 -ugEf <my_genome> <my sorted BAM> | bcftools view -g - > samp.raw.vcf

I got a line in the VCF file that doesn't make sense: It said that there is no variant (ALT=".", AF1="0"), but there are alternative alleles, as can be seen in the DP4 attributes. I also see it clearly in the IGV when viewing the BAM file.

Code:

genome     197469  .       T       .       72      .       DP=666;VDB=0.0469;AF1=0;AC1=0;DP4=258,220,75,110;MQ=33;FQ=-69;PV4=0.0024,1,3.4e-120,1   PL      0

And the weirdest thing is that when I ran the same command on a different sample, the output was ok (both samples contain the same SNP):

Code:

genome     197469  .       T       C       27      .       DP=666;VDB=0.0469;AF1=0.5;AC1=1;DP4=258,220,75,110;MQ=33;FQ=30;PV4=0.0024,1,3.4e-120,1  GT:PL:GQ        0/1:57,0,255:60

Anybody knows what's going on? I've never seen this behavior before with Samtools.

Thanks,
Rachelly.

swbarnes2 · 04-03-2012, 09:17 AM

What does the pileup look like at that locus?

Rachelly · 04-04-2012, 03:12 AM

The pileup of both samples looks exactly the same. See the attached image from IGV.

pileup.png

I found more examples for this behavior in these samples, where they look the same in IGV but their VCF files look different:

Code:

genome     256638  .       C       .       82      .       DP=702;VDB=0.0188;AF1=0;AC1=0;DP4=319,217,94,70;MQ=33;FQ=-79;PV4=0.65,1,1.3e-98,1       PL      0
genome     256638  .       C       A       4.77    .       DP=702;VDB=0.0188;AF1=0.4999;AC1=1;DP4=319,217,94,70;MQ=33;FQ=6.99;PV4=0.65,1,1.3e-98,1 GT:PL:GQ        0/1:33,0,255:33

Code:

genome     217418  .       A       .       15.1    .       DP=940;VDB=0.0028;AF1=0.492;AC1=1;DP4=316,372,116,121;MQ=33;FQ=-12.3;PV4=0.45,2.7e-05,9.7e-135,4.6e-10  PL      15
genome     217418  .       A       T       5.46    .       DP=940;VDB=0.0028;AF1=0.4999;AC1=1;DP4=316,372,116,121;MQ=33;FQ=7.8;PV4=0.45,2.7e-05,9.7e-135,4.6e-10   GT:PL:GQ        0/1:34,0,255:34

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
what is DP4 in VCF (Variant Call Format) and AF1?	gcrdb	Bioinformatics	8	01-20-2015 10:37 AM
count number of A,C,G,T using DP4 in VCF	gcrdb	Bioinformatics	4	03-18-2013 04:28 AM
No reference/alternate (DP4) read depth in VCF?	Tally	Bioinformatics	0	02-20-2012 04:33 PM
How to get list of column in vcf file using Vcf.pm?	jessada	Bioinformatics	0	01-20-2012 08:22 AM
Converting Dindel VCF file to GATK BED file	MolecularToast	Bioinformatics	2	09-24-2011 07:38 PM

04-03-2012, 05:23 AM	#1
Rachelly Member Location: Israel Join Date: Oct 2010 Posts: 37	VCF file - no ALT but DP4 alternative exist Hi all, I'm using samtools to find variants in my resequencing data, and I've noticed something weird: After running the following command: Code: samtools mpileup -L 10000 -d 10000 -ugEf <my_genome> <my sorted BAM> \| bcftools view -g - > samp.raw.vcf I got a line in the VCF file that doesn't make sense: It said that there is no variant (ALT=".", AF1="0"), but there are alternative alleles, as can be seen in the DP4 attributes. I also see it clearly in the IGV when viewing the BAM file. Code: genome 197469 . T . 72 . DP=666;VDB=0.0469;AF1=0;AC1=0;DP4=258,220,75,110;MQ=33;FQ=-69;PV4=0.0024,1,3.4e-120,1 PL 0 And the weirdest thing is that when I ran the same command on a different sample, the output was ok (both samples contain the same SNP): Code: genome 197469 . T C 27 . DP=666;VDB=0.0469;AF1=0.5;AC1=1;DP4=258,220,75,110;MQ=33;FQ=30;PV4=0.0024,1,3.4e-120,1 GT:PL:GQ 0/1:57,0,255:60 Anybody knows what's going on? I've never seen this behavior before with Samtools. Thanks, Rachelly.

04-03-2012, 09:17 AM	#2
swbarnes2 Senior Member Location: San Diego Join Date: May 2008 Posts: 912	What does the pileup look like at that locus?