Linkaged markers not significant

[rui gao]

Hi,

I have been using BLINK to run GWAS on whole genome sequencing data and its working so well. However, I want to ask that Why the markers very nearing the  very significant marker detected by BLINK, are not associate the the phenotype ie P value of linkaged markers are very high than the significant marker?

Thank you so much for any help in advance!!

best,

rui

[Meng Huang]

Hi Rui,

BLINK or FarmCPU's model is a multiple regression model that the genotype data of "possible QTN" were added into the model as covariates to control the inflation, so the -log10(P-value) of the inflated SNPs in the same LD could be very small if there is a "possible QTN" in this LD. 

Here is the FarmCPU paper, which could fully explain this question.

https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1005767

Thanks!

Meng

--
You received this message because you are subscribed to the Google Groups "BLINK Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to blink+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/blink/55476f55-d649-4008-b4ec-4cc5e6668d43n%40googlegroups.com.