|
Post by xyyin666666 on Mar 16, 2016 19:04:24 GMT
Hi there,
I used the GCTA to estimate the genetic relationship matrix in my GWAS imputed + genotype data. In the genetic relationship matrix, I found a lot of negative values. I thought the correlation should be positive. How could it be negative?
Thanks,
|
|
|
Post by Jian Yang on Mar 26, 2016 4:50:28 GMT
If there is no population structure in the sample, the expected value of GRM is zero (Yang et al. 2010 NG). The negative values are expected.
|
|