bizarre bivariate GCTA results

Source Variance SE

V(G)_tr1 0.002403 0.004281

V(G)_tr2 0.013330 0.012013

C(G)_tr12 -0.001414 0.005104

V(e)_tr1 0.070161 0.004486

V(e)_tr2 0.100831 0.012209

C(e)_tr12 0.008501 0.005417

Vp_tr1 0.072564 0.001435

Vp_tr2 0.114161 0.002973

V(G)/Vp_tr1 0.033116 0.058991

V(G)/Vp_tr2 0.116765 0.105100

The estimate of variance explained on the observed scale is transformed to that on the underlying scale:

(Proportion of cases in the sample = 0.078772; User-specified disease prevalence = 0.100000 for disease 1 and = 0.150000 for disease 2)

V(G)/Vp_tr1_L 0.120016 0.213787

V(G)/Vp_tr2_L 0.304784 0.274335

rG -0.249859 0.954170

logL 5871.506

n 8071


Does any know what is going on here? For some reason I am getting a negative value for shared heritability between my two traits?

Thanks