Question: Samtools mpileup: AF1 and DP4 not provided
gravatar for filippo.martignano
2.6 years ago by
filippo.martignano20 wrote:


I'm using this command line:

samtools mpileup SAMPLE.bam -f GENOME.fasta -B -d 1000 -R -Q 20 -l BEDFILE.bed -t ADF,ADR,SP,INFO/AD,INFO/ADF,INFO/ADR -v > OUTPUT.vcf

and the resulting VCF lines look pretty much like this:

chr2 70000005 . T G,<*> 0 . DP=64;ADF=29,0,0;ADR=23,3,0;AD=52,3,0;I16=29,23,0,3,1942,74680,82,2354,3060,183600,180,10800,1072,25682,60,1350;QS=0.958481,0.041519,0;VDB=0.148533;SGB=-0.511536;RPB=0.289223;MQB=0.998456;MQSB=0.970565;BQB=0.0683671;MQ0F=0.03125 PL:SP:ADF:ADR 0,88,255,157,255,255:10:29,0,0:23,3,0

which is different from a "typical" VCF line (example taken from the internet):

chr10 3360 . T C 103 . DP=31;AF1=0.5;CI95=0.5,0.5;DP4=6,3,13,9;MQ=20;FQ=27;PV4=1,0.13,1,0.015 GT:PL:GQ 0/1:133,0,54:57

In particular I was interested in the AF1 and DP4 information which is not provided.

Any ideas?

info af1 mpileup samtools • 1.3k views
ADD COMMENTlink modified 2.6 years ago by Kevin Blighe61k • written 2.6 years ago by filippo.martignano20
gravatar for Kevin Blighe
2.6 years ago by
Kevin Blighe61k
University College London
Kevin Blighe61k wrote:

I run samtools mpileup and then pipe it into bcftools call, which produces AF1 and DP4 each and every time.

Try running it like this (or modify this slightly to suit your needs):

samtools mpileup --redo-BAQ --min-BQ 30 --per-sample-mF --output-tags DP,AD -f GENOME.fasta --BCF My.bam | bcftools call --multiallelic-caller --variants-only -Ov > My.bcf


ADD COMMENTlink written 2.6 years ago by Kevin Blighe61k


Thanks a lot, unfortunately it still doesn't provide such information. Is it possible that it is because I generated a VCF with samtools mpileup instead of a BCF like you did? That doesn't actually make sense to me, but it seems to be the only difference in the pipeline.

ADD REPLYlink written 2.6 years ago by filippo.martignano20

For DNA-seq, I do the alignment with bwa and then it's just a series of samtools and Picard commands before I reach the final variant calling step. Which aligner did you use?

ADD REPLYlink written 2.6 years ago by Kevin Blighe61k

I was using .bam files from Genome In A Bottle project, i belive they are aligned with novalign. I just tried also with TCGA samples that are aligned with BWA (if I am correct) and these are the results:

chr10 47663 . C T 218 . DP=86;ADF=10,19;ADR=0,41;AD=10,60;DPR=10,60;VDB=0.923183;SGB=-0.693147;RPB=0.688282;MQB=0.284284;MQSB=0.443203;BQB=0.211651;MQ0F=0.313953;AC=2;AN=2;DP4=10,0,19,41;MQ=21 GT:PL:DP:DV:SP:DP4:ADF:ADR:AD:DPR 1/1:245,46,0:70:60:43:10,0,19,41:10,19:0,41:10,60:10,60

now I can see the DP4 value, but still AF1 is missing.

ADD REPLYlink written 2.6 years ago by filippo.martignano20

It may be related to program versions, in that case.

I use bwa-0.7.12 (bwa mem), samtools-1.3.1, and bcftools-1.3.1

ADD REPLYlink modified 2.6 years ago • written 2.6 years ago by Kevin Blighe61k
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1833 users visited in the last hour