The terminal nodes are the ones among which you wish to find connections. Each terminal should have a 'penalty' associated with it. The higher the penalty, the more the algorithm will attempt to keep the terminal in the final tree. For details see Huang and Fraenkel. If you have no reason to prefer one node to another, all the penalties can be set to 1. A sample terminal file can be visualized here, and more details on the format are available below.
Note: The algorithm uses separate nodes for mRNAs and the proteins that they encode. Therefore, in the terminal file, each DNA terminal MUST have an 'mrna' tag in the input. In the example above, we have both XYZ and its transcript are nodes.Terminal file template with root:
The user has the option of specifying a root node. In the example below, the inputs are the same as above, but we have indicated that XYZ should be the root. , Next, [terminalset1] tag implies that the terminal list is starting. In the terminal list, name of the terminal node and its corresponding penalty value is given.Terminal file template without root:
You may use our set of protein-protein interactions or provide your own. Currently, SteinerNet provides the following:
The algorithm only has one parameter, which is called ß. This parameter controls the trade-off between the cost of excluding termini from the solution and the cost of including edges. Separate ß values can be used for the protein termini and the mRNA termini.Running your job
The result page includes a basic visualization of the optimum Steiner tree using the Cytoscape Web plug-in . This visualization is provided to give users a quick look before they download output files. Selecting the 'augmented network' gives you the nodes from the prize-collecting Steiner tree with all the edges among these nodes that are present in the interactome, regardless of whether they were included in the tree.Downloads
Networks can be downloaded as Cytoscape-compatible SIF files. In addition, users may download several properties of nodes and edges included in the solution
More detailed information and descriptions about the SteinerNet method and its application to yeast pheromone response are available in ref .
© Copyright The Fraenkel Lab 2012, Massachusetts Institute of Technology. For more information contact firstname.lastname@example.org