Processes 2018, 6,39
network (at the genome-scale level or not): Omix (Omix Visualization GmbH & Co. KG, Lennestadt,
Germany, 2018) [14], MetDraw (freely available at http://www.metdraw.com)[15], MetExploreViz
(available at: http://metexplore.toulouse.inra.fr/metexploreViz/doc/)[16], BiGG (freely available
for academic use at http://bigg.ucsd.edu)[ 3], Fluxviz (Cytoscape pen-source plug-in available
at http://apps.cytoscape.org/apps/fluxviz)[ 17], VisANT (VisANT 5.0 freely available at: http:
//visant.bu.edu)[18], among many others. However, ‘visualization relies mainly on human perceptual
and cognitive capabilities for extracting information’ [19], so many scientists prefer to use their own
network representations of their models which are perfectly fitted to their needs because they are
accustomed to recognizing these classic metabolic pathways and metabolites at a glance. Furthermore,
even on core models one can be confronted with a great (huge in some instances) number of flux data
so that hand-drawing is time consuming. FluxVisualizer is a software that does not seek to compete
with the software mentioned above in drawing metabolic networks from, for instance, an Systems
Biology Markup Language (SBML) file, which most software already does very well. The first aim of
FluxVisualizer is to use a customer’s SVG representation of a metabolic network to simultaneously
visualize reactions and flux values, that is, to automatically draw from the customer’s network
what the biochemist usually draws by hand. The second aim of FluxVisualizer is to automatically
generate a series of pathways on a metabolic network. This is often necessary when dealing with a
list of elementary flux modes (EFMs) or a series of results obtained in Flux Balances Analysis (FBA)
particularly in varying the constraints (Flux Variability Analysis) or with time (dynamic FBA). In these
cases, researchers are faced with tedious time-consuming series of drawings that have to be automated.
FluxVisualizer is an open source software, which offers a simple way to represent fluxes by colour
and/or width on a Scalable Vector Graphics (SVG) image. We chose the SVG format, because it is
widely used and can be built and edited by a variety of programs. The XML structure of the SVG
format makes it easy to access, edit and save in any necessary quality. Furthermore, already existing
software [3,14–18] can output an SVG file of a metabolic network (inputted as XML file for instance).
FluxVisualizer can handle different classical formats of flux distributions (Metatool output files [20],
CellNetAnalyzer export files [21], COPASI export files [22] and FAME export files [23]) and more
generally any CSV or TSV files, so that an Excel file of flux values can be directly represented on
the customer’s SVG image. FluxVisualizer can automatically describe a series of pathways of the
same metabolic network, for instance a series of EFMs, resulting in a set of different SVG files of the
same basic metabolic map. The program provides a graphical user interface (GUI) and an application
programming interface (API) for python3. All functionalities of the program can be used from the API
and the GUI, whereas the API has more possibilities to adapt the output and allows advanced users to
add their own functionalities to the program.
2. Overview of FluxVisualizer
Figure 1 illustrates the idea of the algorithm. Starting from a SVG image of the metabolic network
(Figure 1a) an “Example Flux” is plotted on Figure 1b with the option “Auto width” that automatically
adjusts the width of the arrows to the flux values between two chosen extrema. The pathway with the
flux values is written in place of the “place-here” label in Figure 1a.
145
network (at the genome-scale level or not): Omix (Omix Visualization GmbH & Co. KG, Lennestadt,
Germany, 2018) [14], MetDraw (freely available at http://www.metdraw.com)[15], MetExploreViz
(available at: http://metexplore.toulouse.inra.fr/metexploreViz/doc/)[16], BiGG (freely available
for academic use at http://bigg.ucsd.edu)[ 3], Fluxviz (Cytoscape pen-source plug-in available
at http://apps.cytoscape.org/apps/fluxviz)[ 17], VisANT (VisANT 5.0 freely available at: http:
//visant.bu.edu)[18], among many others. However, ‘visualization relies mainly on human perceptual
and cognitive capabilities for extracting information’ [19], so many scientists prefer to use their own
network representations of their models which are perfectly fitted to their needs because they are
accustomed to recognizing these classic metabolic pathways and metabolites at a glance. Furthermore,
even on core models one can be confronted with a great (huge in some instances) number of flux data
so that hand-drawing is time consuming. FluxVisualizer is a software that does not seek to compete
with the software mentioned above in drawing metabolic networks from, for instance, an Systems
Biology Markup Language (SBML) file, which most software already does very well. The first aim of
FluxVisualizer is to use a customer’s SVG representation of a metabolic network to simultaneously
visualize reactions and flux values, that is, to automatically draw from the customer’s network
what the biochemist usually draws by hand. The second aim of FluxVisualizer is to automatically
generate a series of pathways on a metabolic network. This is often necessary when dealing with a
list of elementary flux modes (EFMs) or a series of results obtained in Flux Balances Analysis (FBA)
particularly in varying the constraints (Flux Variability Analysis) or with time (dynamic FBA). In these
cases, researchers are faced with tedious time-consuming series of drawings that have to be automated.
FluxVisualizer is an open source software, which offers a simple way to represent fluxes by colour
and/or width on a Scalable Vector Graphics (SVG) image. We chose the SVG format, because it is
widely used and can be built and edited by a variety of programs. The XML structure of the SVG
format makes it easy to access, edit and save in any necessary quality. Furthermore, already existing
software [3,14–18] can output an SVG file of a metabolic network (inputted as XML file for instance).
FluxVisualizer can handle different classical formats of flux distributions (Metatool output files [20],
CellNetAnalyzer export files [21], COPASI export files [22] and FAME export files [23]) and more
generally any CSV or TSV files, so that an Excel file of flux values can be directly represented on
the customer’s SVG image. FluxVisualizer can automatically describe a series of pathways of the
same metabolic network, for instance a series of EFMs, resulting in a set of different SVG files of the
same basic metabolic map. The program provides a graphical user interface (GUI) and an application
programming interface (API) for python3. All functionalities of the program can be used from the API
and the GUI, whereas the API has more possibilities to adapt the output and allows advanced users to
add their own functionalities to the program.
2. Overview of FluxVisualizer
Figure 1 illustrates the idea of the algorithm. Starting from a SVG image of the metabolic network
(Figure 1a) an “Example Flux” is plotted on Figure 1b with the option “Auto width” that automatically
adjusts the width of the arrows to the flux values between two chosen extrema. The pathway with the
flux values is written in place of the “place-here” label in Figure 1a.
145