Shareviz is a graph visualization software. The software would take as inputs: specific nodes of data, differently weighted edges between nodes, and identifiers that give nodes color. Then the software will use the LinLog algorithm to position the points in a least energy pattern. Finally, the nodes should display in a java GUI.
The program has been developed by
Itsik Pe'er's Lab of Computational Genetics at Columbia University. It is built in Java 1.6 and tested in the Red Hat Linux environment; the source is distributed here in a jar package under the GPL license.
Installation and Usage
Simply Download the jar file into the directory from which you wish to run it. Put input files into the same directory. Invoke the java command. A new GUI should pop up displaying the clusters and giving buttons for a variety of visual functions.
Command
Use the command java -jar Shareviz.jar
GUI
Edge Length: A slider that can modify the min weight of edges displayed.
Show Labels: A button that will display all the IDs of each node
Reset Layout: A button that returns to the original picture.
Toggle Quality: A button that displays the image in high resolution.
Export: A button that exports the picture to the current computer. To export to C:// drive for example, type in the address line "C:\Sharevizimage.png."
Zoom in and out: Accomplished by clicking and dragging over area to zoom, then pressing "i" or "o" on keyboard.
Rotation: Accomplished by clicking and dragging over area to zoom, then pressing "l" or "r" on keyboard.
For more detailed step to stop description of software, please click here.
Lin Log Model
The clustering mechanism used here is the Ling Log model. The Lin Log model was an energy algorithm designed by Andreas Noack to provide an efficient way to minimize the energy in our force-directed cluster graph. It is this algorithm that we implemented in the creation of our java program. Like all energy models, the Lin Log model has two parts: an energy model and an algorithm to iterate to find the lowest energy. The model is based on the idea that the lowest energy is achieved by minimizing the ratio between the average edge length to average distance to all nodes. It provides the minimal ration of the arithmetic mean of the edges to the geometrical mean of node distances. With each point p, the algorithm aims to achieve for a set a constant c, p’ = c/(|E| arithmeticmean(E, p))*p where E is the edgelength. The LinLog model establishes the cut ratio as the fraction of possible edges between one node and another node. Cutratio(V1, V2) = |E[V1, V2]|/(|V1| · |V2|). Then, the Lin Log model proposes the idea that wherever one cuts a one-dimensional minimum LinLog energy drawing into two nonempty parts, the harmonic mean of the distances between the nodes in the two parts equals the inverse cut ratio.
Logging
The default logging information is set to info.