¹Department of Computer Science and Engineering, Ewha Womans University, Seoul, South Korea.
²Department of Natural Sciences and Mathematics, Wesleyan College, Macon, GA, United States.
³Institute of Bioinformatics, Ewha Womans University, Seoul, Korea & Institute of Bioinformatics, Macrogen Inc., Seoul, Korea.

---

Abstract—Biological pathways can be characterized as networks and grouped as metabolic pathways, gene regulatory networks, gene interaction networks and signal transduction pathways. It is important that edge crossings in biological pathway diagrams are kept to a minimum by strategic placement of vertices for simplification. The basic graph layout technique deals with the problem of positioning the vertices in a way to maximize understandability and usability in a graph. However, when dealing with a very large number of nodes in a global metabolic pathway, strategically positioning vertices is not enough. Understanding the properties of the metabolites and the biological reactions is crucial to pave the way for the formulation of new strategies for further development of automatic layout for global metabolic pathway. In this paper, we provide a statistical analysis of metabolic reactions based on the parsing result of publicly available XML files in KEGG. The analysis leads to a new node-abstracting scheme according to the newly defined concept, ‘relation degree of compound pairs’. The concept would suggest valuable information to software developers for graph-based visualization tools for analyzing networks in cell biology.

Index Terms—Metabolic pathway, parsing, XML, relation degree, drawing algorithm, edge crossing, statistics of metabolites.

[PDF]

Cite: Myungha Jang, Jiyoung Whang, Coleen S. Lewis, Hyun S. Park, "An Interpretation of Biological Metabolites and their Reactions Based on Relation Degree of Compound Pairs in KEGG XML Files," Journal of Software vol. 5, no. 2, pp. 187-194, 2010.