Question: Negative IBD values
3
gravatar for DanielSvensson
3.4 years ago by
Umeå University
DanielSvensson30 wrote:

I'm working with a regional human sample that has been whole-genome sequenced, and right now I'm interested in finding out if I have any relationships between the individuals. For this I'm running KING (http://bioinformatics.oxfordjournals.org/content/26/22/2867).

I've first used Plink to filter out only SNPs with max missingness of 0.1%, then I run King with this command:

king -b $InputBED --homo --showIBD

That is, I'm assuming a homogeneous sample.

I'm getting a lot of negative values for IBD1. Below is an excerpt from the king output file (i've masked the samples):

FID1    ID1     FID2    ID2     N_SNP   IBD0    IBD1    IBD2    Kinship
x       x       x       x       9383369 1.400   -0.566  0.165   -0.0588
x       x       x       x       9383369 1.414   -0.568  0.154   -0.0652
x       x       x       x       9383369 1.390   -0.571  0.181   -0.0522
x       x       x       x       9383369 1.412   -0.572  0.160   -0.0630
x       x       x       x       9383369 1.413   -0.573  0.159   -0.0635
x       x       x       x       9383369 1.426   -0.586  0.161   -0.0662
x       x       x       x       9383369 1.423   -0.591  0.168   -0.0636
x       x       x       x       9383369 1.416   -0.591  0.175   -0.0601
x       x       x       x       9383369 1.428   -0.601  0.173   -0.0638
x       x       x       x       9383369 1.421   -0.610  0.189   -0.0578
x       x       x       x       9383369 1.429   -0.618  0.189   -0.0598
x       x       x       x       9383369 1.471   -0.661  0.190   -0.0703

I'm wondering what the negative values for IBD1 and Kinship mean. My interpretation of what IBD1 is is something along the lines of "for the given pair of individuals, at any given locus, the probability to find exactly 1 common allele is equal to IBD1". But how can it be negative? And how can IBD0 be larger than 1? As you might notice, I'm really confused. Any help is really appreciated!

/Daniel

kinship ibd next-gen king • 1.1k views
ADD COMMENTlink modified 3.2 years ago by cecilia.villacorta50 • written 3.4 years ago by DanielSvensson30

Something I have been wondering about too. Bookmarked!

ADD REPLYlink written 3.4 years ago by MAPK1.4k

Hi Daniel, I'm trying to figure out what panel of SNPs KING interrogates to calculate kinship, but this doesn't seem to be explicitly stated anywhere. Do you know what this panel is?

ADD REPLYlink modified 15 months ago • written 15 months ago by gaelgarcia150
0
gravatar for cecilia.villacorta
3.2 years ago by
Australia
cecilia.villacorta50 wrote:

Hi Daniel,

I just came across the same problem when using KING. I am not sure why I get negative IDB values. Because of these, some pairs of samples have an IBD0 = 2 (you also have IBD0>1), which does not make sense since the maximum value should be 1.

I was wondering if you figured it out?

Regards,

Cecilia

ADD COMMENTlink written 3.2 years ago by cecilia.villacorta50
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1096 users visited in the last hour