Datasets with a high dimensional feature space, advancing statistical methods, and computational efficiency were analyzed to uncover the rules of the circadian rhythms. The aim of the study was to uncover the identity, the dynamic behavior, and the interactions among the components of the circadian clock. Transcriptional profiling has exposed the regulon conferring benefits for circadian biology and bioinformatics. Circadian plant time course gene expression data was examined, this was the prerequisite for Naive Bayes classifiers which were trained and led to expression model with a success rate of up to 87%. The model showed new combinatorial rules, including presence of elements and their frequencies in driving particular phases. Implementation of Genemining V2.3 multipotent algorithm showed the specific combinations of elements responsible for expression patterns, highlighting the role of GATA motifs. State-of-the-art technologies allowed for a model in silico, the first such model was made using time course circadian data.