Tutorial

Example data sets

Download and extract an example data set.

  • Raw PPINs - Contains PPINs of four species (D. melanogaster, H. sapiens, M. musculus, R. norvegicus) extracted from (IntAct, 05/2011) and corresponding node-node similarity files.
    • Use this dataset to walk through this tutorial and create a SMAL MNA around R. norvegicus as the scaffold (review resulting MNA).
  • Illustration - Contains three very small, artificial networks, PNAs and the SMAL MNA.
    • Use this dataset for a quick, manually traceable, illustrative example of the workings of SMAL (review resulting MNA).

Generating PNAs

Open the Generate PNA tab.

  • Select "SMETANA" as the alignment algorithm.
  • Click the "Browse.../Choose File" button under "PPIN1" and select the R. norvegicus PPIN rat.pin from the previously downloaded and extracted example-data folder. Alternatively, drag and drop the file from your file explorer onto the button.
  • Repeat this step under "PPIN2 using the M. musculus PPIN mouse.pin.
  • Next, select mouse-rat.sim under "Node-node similarities".
  • Finally, click the "Generate PNA" button.

You will be redirected to a screen with a link to the directory that will soon contain the pairwise alignment file. Click that link and wait for the alignment to complete. Then download the PNA and save it as mouse-rat.pna.

Repeat these steps another two times to generate the remaining PNAs, using {human.pin, human-rat.sim} and {droso.pin, droso-rat.sim} under "PPIN2" and "Node-node similarity" respectively. rat.pin shall always be used under "PPIN1".

At the end of this section, you will have generated three PNAs (droso-rat.pna, human-rat.pna, mouse-rat.pna). For convenience, these files are also included in the example data set.

Creating a SMAL MNA

At this stage you should have pairwise alignments and corresponding PPINs available.

Open the Create MNA tab.

  • Click the "Browse.../Choose File" button next to "Scaffold PPIN" and select rat.pin from the previously downloaded and extracted example-data folder. Alternatively, drag and drop the file from your file explorer onto the button.
  • Enter "ra" in the adjacent textbox "optional label". If no label is entered, the filename (without extension), "rat" in this case, will be used.
  • Next, select the PNAs and corresponding PPINs (droso-rat.pna, droso.pin, human-rat.pna, human.pin, mouse-rat.pna, mouse.pin) and drop all six files onto the "Browse.../Choose Files" button.
  • Alternatively, to upload and label files separately, use the table.
    • Click "Browse.../Choose File" and select, or drag and drop, droso-rat.pna under "PNA: 'Scaffold <- PPIN'".
    • In the same table-row under "PPIN", select droso.pin.
    • Again, in the same table-row under "optional label", enter "dr".
    • Repeat the last three steps using {human-rat.pna, human.pin, "hu"} and {mouse-rat.pna, mouse.pin, "mo"} on dedicated table-rows.
  • Finally, click the "Create MNA" button.

You will be redirected to a screen with a link to the results directory containing the alignment files as well as links to further actions such as modifying or visualizing the MNA. For convenience, results are readily available and can be reviewed online.

Visualizing an MNA

Note that visualization of large graphs is very resource intensive. Please be patient, cytoscape can become unresponsive for some time when interacting with large networks and take significant time to compute layouts.

Visualize the formerly created MNA by clicking on the "View in cytoscape.js canvas online" link in the SMAL MNA results page.

Alternatively, to visualize any network on this server, open the Visualize MNA tab.

  • In the upper text-field, specify the location of the JavaScript file containing the MNA.
      "examples/smal-smetana-rat-dr_hu_mo_ra/nw.js" for the MNA created in this tutorial based on the "Raw PPINs" data set
      "examples/smal-illustration/nw.js" for the MNA created based on the "Illustration" data set
  • In the lower text-field, enter the number of networks comprising the MNA.
  • Click the "Visualize MNA" button.

As a third alternative, to use the full screen for visualizations, you can open the cytoscape canvas directly and reference networks using URL parameters ("Raw PPINs" example).

The graph will initially be rendered (patience!) in grid layout and reflects the structure of the full scaffold PPIN. Specifically, the node labels are the proteins of the scaffold PPIN. Note that for large graphs labels will not be visible initially as the text would be too small to render.

  • Filters allow to restrict the current collection (visible/rendered subnetwork) to nodes and edges that are conserved in a certain number of species.
  • Nodes can be dragged across the canvas.
  • In the "search" text field on the upper-right, enter "P62161". The graph will pan and zoom in on the specified vertex.
  • Clicking the node opens a dialog containing information about the alignment cluster constitution and functionality for graph manipulation (see more details below). For now, click on "Filter neighborhood". This will remove all but the selected protein and its direct neighbors.
  • Select the "circle" layout to rearrange the new selection of nodes on the canvas.
  • Click on the neighboring node "P30543" and then click on "Add neighborhood (1 step)".
  • Use the "zoom" buttons to zoom in or out.
  • Select the "cose" layout and click "Apply layout".
  • Use the mouse to pan and zoom onto the graph.
  • Click on node "Q62968" and "Add neighborhood (1 step)".
  • You can go back one step by clicking on the "Restore previous selection" button.
  • In the dialog of node "Q62968", click on the UniProt search link.
  • Review the node in UniProt and make note of some of its known interactors.
  • Seach for the listed interactors "Q9Z336 Q63425 P05943" and they will be added to the selection of visible nodes.
  • Redraw the layout and zoom/pan onto the new selection.
  • Note that node "P05943" has only three colors (grey, red and blue). Those colors match the species that are part of the node alignment cluster. "Q63425" contains four colors since nodes from all four species part of the MNA are aligned. In addition to the number of colors, node size indicates the number of aligned species as well (the more, the larger).
  • Edges are colored according to which species have conserved interactions to the scaffold interaction in grey as well. As
  • nodes, the weight of the scaffold interaction is indicative of the number of species contributing induces edges.
  • Finally, the dialog also offers functionality to remove nodes.