Table 6.
Recessive Model | Genotype effects (3 covariates)1 |
Genotype effects (4 covariates)2 |
Genotype by diagnosis interaction3 |
---|---|---|---|
Total expansion (cubic mm) |
4,5F-ratio=5.710 p=0.017 |
F-ratio=6.198 p=0.013 |
F-ratio=0.134 p=0.715 |
Left expansion (cubic mm) |
F-ratio=6.230 p=0.013 |
F-ratio=5.817 p=0.016 |
F-ratio=0.075 p=0.784 |
Right expansion (cubic mm) |
F-ratio=4.460 p=0.035 |
F-ratio=4.684 p=0.031 |
F-ratio=0.123 p=0.726 |
Additive Model |
Genotype effects (3 covariates)1 |
Genotype effects (4 covariates)2 |
Genotype by Diagnosis interaction3 |
Total expansion (cubic mm) |
F-ratio=3.330 p=0.037 |
F-ratio=3.873 p=0.021 |
F-ratio=1.142 p=0.320 |
Left expansion (cubic mm) |
F-ratio=3.384 p=0.035 |
F-ratio=3.630 p=0.027 |
F-ratio=0.390 p=0.677 |
Right expansion (cubic mm) |
F-ratio=3.407 p=0.034 |
F-ratio=3.744 p=0.024 |
F-ratio=2.216 p=0.110 |
rs6347 genotype (top: recessive model; bottom: additive model) was used to predict variations in ventricular expansion, with age, sex, and diagnosis regressed out.
rs6347 genotype (top: recessive model; bottom: additive model) was used to predict variations in ventricular expansion, with age, sex, diagnosis, and ApoE status regressed out.
Significance of the genotype by diagnosis interaction term using the following equation: expansion measure = constant + genotype + age + sex + diagnosis + genotype*diagnosis.
In multiple regressions, the F-ratio is used to test the hypothesis that the slopes of the regression lines are 0. The F is large when the independent variable helps to explain the variation in the dependent variable, independently of the other explanatory variables that are regressed out. For instance, here we reject the hypothesis that the slope of the regression line is 0 (F-ratio=5.710, p=0.017), meaning that there is a significant linear relation between rs6347 genotype and total ventricular expansion, independent of age, sex, and diagnosis.
Bold font indicates significant results (p<0.05), and regular font indicates results that did not reach statistical significance.