This course teaches one how to use the statistical language R to specify, learning, and interpret Structural Equation Models (SEM) with a causal inference-based approach. SEM is an important area of discussion about the causal mechanism of many interconnected variables. Classically associated with psychology, sociology and economics, they are now also widespread in biology (in network-based interpretation of high-throughput data), medicine (in neuroimaging) and bioinformatics (in Bayesian Networks). SEM amalgamates regression analysis, path analysis and factor analysis. Using the language of graphical models, SEMs allow to specify a flexible class of models defined on hypotheses formulated a priori (confirmatory models) and modifiable on the basis of the structure present in high dimensionality data (learning models). The course comprises of a mixture of short lectures and R sessions on the basic theory behind SEM and Structure Learning, teaching via real data set examples, which participants can follow, and exercises to practice the skills just learned.