Figure 1

Workflow diagram showing the model construction used to create the Vascular Inflammatory Processes Network (V-IPN). (A) The V-IPN is comprised of two primary components: a literature model (blue cylinder) and an RCR-based integration of gene expression vascular datasets (green cylinder). The literature model was constructed from causal connections within well-defined biological contexts and boundaries using published scientific literature. The content of the literature model was augmented with nodes derived from RCR analysis of publicly available vascular inflammation-relevant datasets. The final V-IPN model (orange cylinder) resulted from a comprehensive review of the integrated model. The final model was then evaluated using additional, independent, gene expression datasets to generate HYPs that were mapped to the V-IPN (purple cylinder). (B) Assembly of scientific knowledge into a model. Published knowledge is described in Biological Expression Language (BEL) and then integrated and assembled in a model as nodes connected by causal relationships (edges). This assembled model/network serves as a substrate for RCR and HYP generation. Each causal relation in the network is based on one or more scientific findings. (C) Mapping of differential measurements to a HYP. A HYP is evaluated as an explanation for the observed changes in gene expression data. Differential measurements in the dataset (RNA abundance of genes A-F expressed as discrete State Changes) are associated with the corresponding downstream nodes of the HYP.