SEM: Structural Equation Modeling in Molecular Biology

Abstract

Structural equation modeling (SEM) is a second-generation multivariate method to estimate the causal interactions in a set of variables and includes, as special cases, several statistical methods (regression analysis, path analysis, and confirmatory factor analysis). This review focuses on all of the main SEM models and various methods used to optimize the model parameters. Representative examples are discussed to illustrate SEM application in molecular biology, including modeling of biochemical processes, relationships between genetic markers and diseases, and interactions within gene networks.