We've moved!
This site is now read-only. You can find our new documentation site and support forum for posting questions here.
Be sure to read our welcome blog!

Genotype Concordance: Logic behind defined values

Hi there,

I am currently using GenotypeConcordance through picard to generate a summary of the concordance between two different sets. One set, a plasma sample, is regarded as the truth, and the other, a urine sample, is regarded as the call/eval sample. VCF files were generated and then filtered for a Quality of 30 using BCFtools. I had no large trouble generating the summary, but I am having a bit of trouble understanding some of the metrics listed in the summary table/text.

For my results, I am finding the non-reference genotype concordance and the reference genotype concordance to be the same value. When looking at the summary/definition given in the picard metric definitions the only difference seems to be in the variants. Is there something more that the reference genotype concordance should normally have that would change their values?

As for another question about the logic behind the calculations, if my call (urine) sample does not contain information on a variant defined by my truth (Plasma) sample, is this considered a False Negative or is this variant removed from analysis?


Best Answer


Sign In or Register to comment.