GL fields in output VCF don't sum to 1 about angsd HOT 2 CLOSED

Ok, that is a documentation bug in the header of the vcf. I’ve changed this in the latest commit on github.

It is loglikeratio to the most likely, and scaled as log10.
Let me know if you find other issues with the vcf. I dont think this is the most used part of angsd.

On 22 Sep 2015, at 01:47, Evan McCartney-Melstad [email protected] wrote:

Hello,
I've been trying to output a VCF file to look at IBD between samples, and I was curious if the GL's in the VCF file are scaled in any way. It says they have been scaled to add up to 1 in the header of the VCF, but they don't (the GP fields do sum to 1).

Here's the command I used:
angsd0.902/angsd -bam bamlist.txt -out OUTPUT -uniqueOnly 1 -only_proper_pairs 1 -remove_bads 1 -doCounts 1 -nThreads 27 -doMaf 1 -doMajorMinor 2 -GL 2 -minMapQ 20 -minQ 30 -dovcf 1 -doPost 2

And here's what my VCF looks like:

##fileformat=VCFv4.2(angsd version)
##FORMAT=
##FORMAT=
##FORMAT=
##FORMAT=
#CHROM POS ID REF ALT QUAL FILTER INFO FORMAT ind0 ind1
scaffold_0 1616 . T A . PASS . GP:GL 0.666631,0.333333,0.000035:0.000000,-0.301007,-4.277052 0.666578,0.333333,0.000088:0.000000,-0.300972,-3.876948

Thanks very much for any insight!
Evan

—
Reply to this email directly or view it on GitHub #18.

from angsd.

Ok, I'm closing this issue. Feel free to reopen if needed

from angsd.