To run SurvNet, you have three options:

1. Analyze Your Data, which is to identify network-based biomarkers in user-provided datasets
2. Analyze Preloaded Data, which enables users to explore preloaded, multidimensional TCGA molecular data.
3. Analyze a set of sample data (no user-provided input files are required).

To analyze your data, you need to upload three input files:

1. a network data file
2. a molecular profiling data file
3. a survival data file

Sample datafiles are available here.

To analyze preloaded data, you need to make a selection from each of the three pulldown menus.

You also need to specify a "searching distance" parameter. SurvNet will automatically start with each gene (or protein) node as the seed and search the optimal subnetworks within this distance. SurvNet uses the same network searching algorithm described in Chuang et al. (Molecular Systems Biology 2007;3:140) and Jia et al. (Bioinformatics 2011;27:1). The larger this parameter is, the longer the computation will take.

Once you upload/select the input files and specify the required parameters, click the Submit button to run the job!

Try out a set of sample data files (no user-provided input files are required).

1. Summary page
   After submitting a job, SurvNet will first check the data quality. If the input files are good, it will generate a data summary indicating the gene number, the sample number and so on.
   
2. Intermediate page
   If the submitted job takes some time, an intermediate page will be generated indicating the job is still running. You can check the result by refreshing this page after a few minutes (see red box 1). Alternatively, the intermediate page also contains a weblink, and you can bookmark the link (see red box 2) and access the results later (the results will be maintained for two weeks).
   
3. Final output page
   In the final output, you will see the results in the table format. You can download the table and the DOT format subnetwork file (see red box 1) which can be visualized by GraphViz. The identified top subnetworks rank with the network p-values (from low to high). The Network p-value is calculated based on the Cox proportional hazards model and network properties, which quantifies how significantly nodes in a sub-network correlate with the observed patient survival data. The higher the score is, the more significant the network correlates with the patients' survival. The Cox p-value is the p-value derived from multi-variable Cox proportional hazards regression model for evaluating the utility of subnetworks. The Gene ID indicates the seed node for each sub-network; the #Nodes indicates how many genes (or proteins) are in the network; the #Edges indicates how many interactions are in the network; and you can visualize the network by clicking the Graph button (see red box 2). The identified subnetworks can be filtered by two parameters (see red box 3). There will be no Graph button (see red box 4) if the sub-network is trivial (only including a single node).
   
   
4. Network visualization page
   After clicking the Graph button in the final output page, you can visualize a network graph in a user friendly interface (it may take a few seconds to load a network graph). Using the buttons in the red box 1, you can (i) adjust the location of a network; and (ii) zoom in/out of a network. You can go back to the final output page by clicking the Back to results list button in the red box 2.