Graph-based Temporal Mining of Metabolic Pathways with Microarray Data

We propose a dynamic graph-based relational learning approach using graph-rewriting rules to analyze how biological networks change over time. The analysis of dynamic biological networks is necessary to understand life at the systemlevel, because biological networks continuously change their structures and properties while an organism performs various biological activities to promote reproduction and sustain our lives. Most current graph-based data mining approaches overlook dynamic features of biological networks, because they are focused on only static graphs. First, we generate a dynamic graph, which is a sequence of graphs representing biological networks changing over time. Then, our approach discovers graph rewriting rules, which show how to replace subgraphs, between two sequential graphs. These rewriting rules describe the structural difference between two graphs, and describe how the graphs in the dynamic graph change over time. Temporal relational patterns discovered in dynamic graphs representing synthetic networks and metabolic pathways show that our approach enables the discovery of dynamic patterns in biological networks.