Mx scripts library: structural equation modeling scripts for twin and family data.

Structural equation modeling (SEM) provides a flexible tool to carry out genetic analyses of family and twin data. The basic model which decomposes the variance between and within families for a particular trait into genetic and non-genetic components can be generalized to multivariate and/or longitudinal data, incorporate sex differences in parameter estimates, and model the effects of measured environment, candidate genes or DNA marker data. We introduce a web-based library (http://www.psy.vu.nl/mxbib) of scripts for uni- and multivariate genetic epidemiological analyses, as well as for linkage and genetic association tests. The scripts are written to be used with the freely available software package Mx and provide a flexible and uniform approach to the analysis of data from relatives.