Skip to: Site menu | Main content


DNA microarrays are used to measure the expression levels of thousands of genes simultaneously. In a time series experiment, the gene expressions are measured as a function of time. GENeVis is an application for integrated visualization of genome expression and network dynamics in both regulatory networks and metabolic pathways. Integration of these two levels of cellular processes is necessary, since it provides the link between the measurements at the transcriptional level (gene expression levels approximated from microarray data) and the phenotype (the observable characteristics of an organism) at the functional and behavioral level. The integration requires visualization approaches besides traditional clustering and statistical analysis methods. Our application can (i) visualize the data from time series experiments in the context of a regulatory network and KEGG metabolic pathways; (ii) identify and visualize active regulatory subnetworks from the gene expression data; (iii) perform a statistical test to identify and subsequently visualize pathways that are affected by differentially expressed genes.

GENeVis has features that are currently lacking in existing tools, such as mapping of expression value and corresponding p-value (or other statistic) to a single visual attribute, multiple time point visualization, and visual comparison of multiple time series in one view. Various interaction mechanisms, such as panning, zooming, regulator and target highlighting, data selection, and tooltips support data analysis and exploration. Subnetworks can be studied in detail in a separate view that shows the network context, expression data plots, and tables containing the raw expression data.


GENeVis considers metabolic pathways as disjoint processes. Most metabolites (i.e. compounds or reactions), however, are shared among several pathways. Therefore, to investigate an organism's metabolism, it is relevant to study all the pathways simultaneously. Therefore, we initiated work on GENeVis II, which enables visualization of the whole metabolic network. It makes use of a layout algorithm called MetaViz, which has been designed specifically for metabolic networks. In addition to visualization of data from time series experiments in the context of a regulatory network and a metabolic network, GENeVis II supports identification and visualization of affected subnetworks based on the expression data. The nodes of a subnetwork are highlighted in both the regulatory network and metabolic network. This approach can reveal quickly whether two independent regulatory subnetworks are connected at the metabolic level, or, conversely, whether two independent metabolic subnetworks are connected at the regulatory level.





November 2009 update: The bget method of the KEGG API has changed, and we are adapting the tool. The updated version should be available for download shortly.

Here is a movie (23 MB) that shows network activity in E. coli. Read the details in the Computer Graphics Forum paper (available above).

GENeVis images

Main application window after loading B. subtilis genome, its transcription network, a two time series containing gene expression ratios and gene expression levels, respectively. See paper for details.

The network window shows the transcription network and gene expression time series. The data shown correspond to a part of the gene regulatory network of the bacterium B. subtilis, and an expression ratio time series of four time points. Expression ratio is mapped to color. The height of a time point box reflects the confidence: taller means higher confidence. Neighbor highlighting (of the gene ccpA) assists the user in understanding network structure. The interaction type is mapped to a color: red for inhibition, green for activation, and grey for other cases.

Visualization of an expression ratio time series of four time points in the context of a part of the purine metabolism pathway in B. subtilis.

Part of the active network in B. subtilis at the last time point of a four-point time series. Active nodes and edges are drawn opaque and highlighted.

Active E. coli subnetwork at time point 6; the expression ratios determine the gene box fill colors. The LacI regulon is shown enlarged in the top right corner. There, a standard fill color is used, and the whole time series is shown with the expression glyphs colored by ratio. The genes lacI, lacA, lacY, and lacZ are all upregulated, as is indicated by red-colored expression glyphs, which reveals that the organism adjusts to growing on lactose.

GENeVis II images

Active E. coli subnetwork at time point 6. The hotspot containing lacZ has been selected in the transcription network (left). The red hull indicates the region in which the hotspot members lie. The same members are highlighted in the metabolic network (right). [click for larger image]

Still at time point 6, the pathway Lactose degradation II has been selected in the metabolic network. The small red region near the center highlights the pathway members. The same members are highlighted in the transcription network on the left. [click for larger image]

Close-up view of the lactose degradation pathway.