|
Post by aleixarnau on Oct 22, 2015 17:48:08 GMT
Hi all,
I'm running some analysis using a bivariate trait code as: 0, 1, NA Here is an example of the command I'm using:
gcta64 --grm geneset2_GS14K_original --reml --pheno phenoGCTA_FID_IID_MDD --prevalence 0.122 --covar covarGCTA_FID_IID_Sex_Age --qcovar fullSetGS14K.eigenvec --out geneset2_GS14K_original --thread-num 20
However, --prevalence is not working, it is recognised by the software but doesn't provide V(G)/Vp_L. And I am interested on h2 ( V(G)/Vp ) on the underlying scale. The output I get looks like that:
Source Variance SE V(G) 0.000068 0.001785 V(e) 0.203593 0.003847 Vp 0.203661 0.003424 V(G)/Vp 0.000334 0.008765 logL 1942.415 logL0 1942.415 LRT 0.001 df 1 Pval 0.4847 n 7175
Do you know or have an idea why it doesn't work?
Tanks in advance,
|
|
|
Post by Jian Yang on Oct 23, 2015 1:18:38 GMT
It's likely that the phenotype was not recognised a case-control study. You might check the log file.
|
|