Effective Visualization for Investigating Elementary Flux Modes in Genome-Scale Metabolic Models

Elementary Flux Modes (EFMs) are an indispensable tool for constraint-based modeling and metabolic network analysis¹. EFMs allow exploration of cellular metabolism beyond conventional pathway definitions and integration with high-throughput data. Since the field has been driving towards improving algorithms to enumerate all (or subsets) of EFMs, their visualization has been limited to mapping only the flux data in standalone tools. However, flux mapping is just one component of systems studies, where several molecular datatypes are combined. There is a need for an efficient visual analysis approach that complements current efforts integrating omics data with genome-scale metabolic models. In this study, we developed a simple, efficient and MATLAB-based workflow for graphically visualizing EFMs as a network of reactions, metabolites and genes. Our flexible workflow seamlessly integrates COBRA² or RAVEN³ with the open-source tool, Cytoscape⁴. Cytoscape not only enables data visualization and advanced network analysis, but also network extension with other molecular interaction datatypes, all within a single framework. Once processed through our workflow, SBGN layout is automatically applied on the visualized network requiring little-to-no manual effort by the user. Furthermore, combining existing tools avoids additional installation of flux-mapping tools, used previously, thus making it both user-friendly and time-saving. We illustrate our workflow using a subset of EFMs generated from both small and large models and map gene expression data on the visualized EFMs, thereby, demonstrating that efficient, integrative visualization can allow for a better understanding of biological processes.

References: [1] Klamt, S. et al., DOI:10.1371/journal.pcbi.1005409. [2] Schellenberger J. et al., DOI:10.1038/nprot.2011.308. [3] Agren, R. et al., DOI:10.1371/journal.pcbi.1002980. [4] Shannon, P. et al., DOI:10.1101/gr.1239303.