TEMPO (TEmporal Modeling of Pathway Outliers) is a pathway-based outlier detection approach for finding pathways showing significant changes in temporal expression patterns across conditions. Given a gene expression data set where each sample is characterized by an age or time point as well as a phenotype (e.g. control or disease), and a collection of gene sets or pathways, TEMPO ranks each pathway by a score that characterizes how well a partial least squares regression (PLSR) model can predict age as a function of gene expression in the controls and how poorly that same model performs in the disease. TEMPO v1.0.3 is described in Pietras (2018) <doi:10.1145/3233547.3233559>.
