Discrepancy in DP value in vcf file and reads visualization in IGV

[vaibhavvsk]

A sample variant from my vcf file :

I 27000 . A G 37.88 . AC=2;AF=1.00;AN=2;DP=51;Dels=0.00;FS=0.000;HRun=1;HaplotypeScore=0.0000;MQ=6.53;MQ0=49;QD=0.74 GT:ADP:GQ:PL 1/1:37,14:51:6.01:69,6,0

which gives DP=51 but while visualizing the variant on IGV i'm getting
Total Count as 56. Seems like some reads have been disqualified for calling variant and the number is lesser in vcf than the bam file.

I need to understand this discrepancy. I would like to view same counts while in vcf and in viewer.

[m_two]

What software produced the output?

Does the meta-information line match this?
##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Read Depth">

If the vcf metainfo matches the line below refer to documentation at https://www.broadinstitute.org/gatk

##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Approximate read depth (reads with MQ=255 or with bad mates are filtered)">

If you are using varscan see:

http://varscan.sourceforge.net/support-faq.html#read-counts-different

[dpryan]

It's also quite likely that whatever made the VCF file was accounting for overlapping paired-end reads, whereas IGV doesn't.

[vaibhavvsk]

What software produced the output?

Does the meta-information line match this?
##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Read Depth">

If the vcf metainfo matches the line below refer to documentation at https://www.broadinstitute.org/gatk

##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Approximate read depth (reads with MQ=255 or with bad mates are filtered)">

If you are using varscan see:
http://varscan.sourceforge.net/suppo...unts-different

As per your quote above my vcf file header shows:

My vcf file has been generated through UnifiedGenotyper.

##INFO=<ID=DP,Number=1,Type=Integer,Description="Filtered Depth">

[Jessica_L]

I see something similar in my outputs from Illumina's somatic variant caller. My relevant header line is this:

##INFO=<ID=DP,Number=1,Type=Integer,Description="Total Depth">

and my allele depth values don't add up to the total, although it's usually only off by fewer than 10 reads.