Post by dianed on Jun 22, 2018 16:37:29 GMT
Hi,
I estimated the heritability for a specific disease. As the result is surprising, I would like to confirm it by permuting the phenotype (cases and controls). So I switched the phenotypes: the cases became the controls and the controls became the cases in the .fam and the phenotype file. I created the grm matrix and I ran reml analysis with the option --prevalence as it is a case/control study. I thought that I will obtain a very small heritability with the permutation of the phenotypes but I get the same heritability estimation. Is there something that I didn't understand or that I did wrong?
Thanks
I estimated the heritability for a specific disease. As the result is surprising, I would like to confirm it by permuting the phenotype (cases and controls). So I switched the phenotypes: the cases became the controls and the controls became the cases in the .fam and the phenotype file. I created the grm matrix and I ran reml analysis with the option --prevalence as it is a case/control study. I thought that I will obtain a very small heritability with the permutation of the phenotypes but I get the same heritability estimation. Is there something that I didn't understand or that I did wrong?
Thanks
