Fig. 1

Schematic TPMS approach used to evaluate the Still’s disease treatments efficacy and their MoA. TPMS is based on systems biology-based models. TPMS encompass five steps: (i) the learning process of the protein-protein interaction (PPI) human network based on training and validation using known information stored in the truth table; this learning is performed with artificial intelligence techniques to construct accurate mathematical models that simulate the behavior of human physiology through two main strategies (ANNs and sampling-based methods). (ii) The molecular characterization of Still’s disease and drugs through dedicated bibliographical revision, from which a Still’s disease interactome can be constructed using the PPI human network. (iii) ANN evaluation of drugs efficacy over Still’s disease definition. (iv) The construction of specific MoA models for Still’s disease and drugs. (v) The validation of the MoA models with available gene expression data from sJIA patients. Expression data enrichment analysis was also used to be compared with the Still’s disease interactome to evaluate the validity of the bibliography-based Still’s disease molecular characterization