Skip to main content
Fig. 1 | BioData Mining

Fig. 1

From: Testing the assumptions of parametric linear models: the need for biological data mining in disciplines such as human genetics

Fig. 1

The top panels show the trait relationships between sib pairs simulated using additive (a), epistasis (b), and additive plus epistasis (c) models. Least squares fit regression lines and the correlation coefficient are shown for each. The bottom panels show the relationship between trait correlations estimated from sib, half-sib, and cousin pairs plotted against the expectation of their genetic relatedness. The panels d-f correspond to the genetic models labeled in a-c. Note that the linear trends in c and f are present despite more than 60% of the heritability being due to epistasis as a nonlinear genetic effect

Back to article page

BioData Mining

ISSN: 1756-0381