How Bipartite Network Visualizations Complement Ingenuity Pathway Analysis: A Case Study in Methylation Related to Preterm Births

Although bipartite networks have been effective in identifying how significant biomarkers are associated to different subsets of patients, little is known about how domain experts use such patterns to infer biological pathways. Here we present a case study to elucidate how patterns from a bipartite network visualization were used in conjunction with Ingenuity Pathway Analysis (IPA) to infer pathways related to methylation in spontaneous preterm birth. The results suggest that because the network made explicit the inverse symmetrical relationship among cases/controls and methylation sites, the visualization helped to rapidly integrate pathways related to cases and to controls into a unified pathway hypothesis for spontaneous preterm. These results elucidate the complementary role that bipartite networks can play in inferring biological pathways from databases such as IPA.