Table 1 |
|||||
|
RedeR in the context of gold-standard network visualization software and R |
|||||
|
RedeR |
Cytoscape |
Graphviz |
igraph |
graph |
|
|
|
|||||
|
Software design |
|||||
|
Hierarchical data structurea |
Yesb |
No |
Yesb |
No |
No |
|
Data abstractiona |
Modularb |
Flat |
Layeredb |
Flat |
Flat |
|
Data encapsulationa |
Yes |
Yes |
No |
No |
No |
|
Core engine |
Java |
Java |
DOT |
C |
R |
|
R interface |
R <-> Java |
R -> Java |
R <- DOT |
R <- C |
Pure R |
|
Deployment to R |
Embedded |
External |
External |
Embedded |
Embedded |
|
Plugin coding language |
R |
Java |
DOT languagec |
Cd |
- |
|
Selected features |
|||||
|
Scalability on nested networkse |
Yes |
Nof |
Yes |
Nog |
Nog |
|
Interactive graph handling |
Yes |
Yes |
No |
Partiallyh |
No |
|
Dynamic layouts |
Yesi |
No |
No |
No |
No |
|
Comparison across multiple nested networks |
Yes |
No |
No |
No |
No |
|
Panelsj |
Yes |
No |
No |
No |
No |
|
|
|||||
|
aFor further definitions, please see the Implementation section and Figure 2. bIn RedeR, the hierarchy is encapsulated, so users deal with methods. In Graphviz users implement methods by themselves in DOT language. This difference has an important effect on the ability of the software to deal with modular structures, such as recycling data objects used in subnetworks. cThe plugin infrastructure is available for Graphviz. dProvides a C library to be used in third-party applications. eAble to accommodate increasing amounts of nested objects. fIn Cytoscape, nested networks are represented as images inside nodes. For each nested network one image is required. Images are not scalable, so the hierarchy cannot be extended to other levels. gIn these applications, modules can be represented by layering images, which is essentially a drawing process. hAccording to the authors, the interactive features are not very well developed at this stage. iIn RedeR, the dynamic layout is also extensible to the nested structures. jThis feature provides support for multiple panels in the same graph, representing subnetworks, and so on. |
|||||
|
Castro et al. Genome Biology 2012 13:R29 doi:10.1186/gb-2012-13-4-r29 |
|||||