Protein Interaction Network Visualizer

A short tutorial

Follow these instructions to make use of some of the tools that PINV provides. If you have any problem or find a bug please report it to https://groups.google.com/group/pinv_biosual. 

The example explores some proteins that are known to be associated with tumor suppression in humans.

  1. On the PINV web page (http://biosual.cbio.uct.ac.za/pinv) choose the dataset called Homosapiens. Once you have clicked on it, a random selection of interactions will be displayed.

  1. On the top-left corner of the app, there is a text area to select the proteins of interest using UniProt accession numbers. Input the following protein’s accessions and press Search:  O15119, Q13227 and Q9UQL6. The graphic should now look like the image below.
    NB: You can add more than one protein at a time by separating them with commas.

TO DO: Use direct manipulation of the graphic to move your proteins away from the border.

  1. Using Pinv you can display different labels on the proteins of interest. For this example, we will display the gene names with more than one interaction.
    On the top right of the screen you will find a link to add a new rule “
    +Add Rule”. Click on it.Rules in PINV have two main parts: an action and a filter. In this case, the action is to “Show Label”. This will display a new option selector to choose which feature to display as label. Select “gene-name”. For the filter you must first choose between “proteins” or “interactions”. Select the first one, as our goal is to display the label at the side of the nodes. This selection can now be filtered. The default is “no filter” meaning that all proteins are included, we however want to select the “number of interactions” option. This kind of filter requires the use of a comparison operator between the options: equal to (==), greater than (>), less than (<), greater than or equal to (>=) or less than or equal to (<=). Fill in the rest of the form and click on APPLY.
    Now the 3 proteins with more than one interaction have the corresponding label, as in the figure above.

  1. You might have noticed in step 2 that under the text area there is a mode selector. Back in step 2 we didn’t change it because we were looking for a Normal query execution.

    Now we are going to query another protein but in
    Explicit mode, which means that the protein to be queried is going to be drawn in the graphic, including only the  interactions with proteins that are already in the graphic. Proceed to the selection of Explicit mode and now query the protein Q9Y618. The proteins are now interconnected thanks to this new addition. Also notice that the rules apply to any new addition to the graphic, and now all the connecting proteins have a label too. Your graphic should look like this now:

TO DO: Create a rule to paint the border of any interaction with score higher than 0.7 green. Add another rule to paint the border of any interaction with score lower than 0.5 red.

  1. Download the file from this URL (http://biosual.cbio.uct.ac.za/example.txt). This is a tab separated file with made-up values for some of the proteins in the graphic and its content is as follows:

                Q9UQL6        0.46        0.90

Q13227        0.50        0.70

O15119        0.53        1.00

Q9Y618        0.44        0.90

Q09472        0.57        0.49

Q14353        0.57        0.49

O75376        0.41        0.01

PINV allows the use of a file in this format to color the proteins in the graphic. To load the file click on the link under the rules component.


Once you select the file, PINV will ask you to select which column to use to color the nodes. In the example, the first column corresponds to a healthy sample and in this case it means that all the expression levels are around 0.5. The second column is supposed to be from a tumor infected sample, and therefore some of the values are distant from the hypothetical desired value. Select the second column.

Notice that only 5 rows are displayed here. The purpose of this table is only to help select a column and not to explore the data.

Subsequently, PINV will ask you to choose the corresponding colors of the maximum, minimum and center point. PINV will interpolate colors between these values.

Input similar values to the following figure:

The current figure should appear as follows. Remember you can use the mouse scroll wheel to zoom in/out. Also notice the SVG and PNG buttons on the button-left corner of the figure - you can take a snapshot of your visualization at any time with them.