topleftcorner logo
Home dash Download dash Help dash Contact
toprightcorner
 

basic search result example - click to resize Search

In the Basic Search, the user can select a kinase and/or a substrate. If you select only one, keep the other as a wildcard (*).

The tabular results of predicted interactions can be directly downloaded, in "tab-separated-values" (.tsv) format, as a spreadsheet or in Cytoscape "sif" format.

There is a background color convention as shown in the screen-shot example on the left. Looking down any column, any time the Kinase or the Substrate changes from the preceding one, the background color changes to accentuate the difference.

In the example on the left (click image to resize) the search was performed for the substrate JUN and all the kinases (*). One can see the background color changing as the Kinase changes. shortest path example - click to resize Since the table was designed to be as compact as possible, some of the additional information (e.g., the description of the kinase) is revealed only when you "mouse-over" the corresponding field value.

If you checkmark a number of rows and press the [Show STRING context] button, a view will pop-up containing the corresponding kinases, substrates and their known associations in the context of STRING.

The column named Source opens a hyperlink and shows the default phosphorylation sites (from pELM and/or PhosphoSite) used for Kinase prediction for the particular Substrate that was selected.
The last column is a link that shows the Shortest Path in STRING graph, between the Kinase and the Substrate for the particular row that was selected (see image example to the right).


advanced search result example - click to resize Advanced Search

The Advanced Search was developed to extend searches beyond the basic match of substrate to kinase and allows the user to filter their search based on multiple criteria.

The user can enter, as filtering criteria, any value fragment or value pattern for any of the fields. For example, a pattern for the Peptide field might be something like %DLF% where the percentage (%) character is used as a wildcard. For numerical fields, such as scores and position, numerical comparison operator must be used (e.g., > or = ). If more than one filter is selected, the results are considered together ("AND" statement). For "OR" statements, download the results as a text spreadsheet and merge. Click image to the left for an example. shortest path example - click to resize

Similar to the basic search case, if you checkmark a number of rows and press the [Show STRING context] button, a view will pop-up containing the corresponding kinases, substrates and their known associations in the context of STRING.

The last column is a link that shows the Shortest Path in STRING graph, between the Kinase and the Substrate for the particular row that was selected (see image example to the right).

The same background color convention applies. Looking down any column, any time the Substrate or the Kinase changes from the preceding one, the background color changes to accentuate the difference.


map example - click to resize Map View

The Map shows a comprehensive view of all kinases and substrates and their predicted interactions.

Kinases are shown in orange and Substrates in blue. Each protein is shaded according to the number of its predicted matches. A darker background indicates a higher number of matches.

Click on any protein to select it. Both that protein and its predicted matches (substrates for a kinase and kinases for a substrate) will become highlighted in green. Clicking on a highlighted protein will de-select itself and its specific predicted matches. In other words, de-selecting a kinase for example will deselect only the substrates that are specific to that kinase and not the ones that are common to other already selected kinases. An example can be seen by clicking the left image.

A gene that lacks an official gene symbol will be displayed by its ENSEMBL Ids. In these cases, a description of the gene can by seen by placing the mouse-over the ID. map view in String example - click to resize

Highlighted proteins can be seen in a "String context" (the network of associated proteins) by clicking on the [Show String context] button. An additional bar-graph in the String view shows the probability of phosphorylation of each selected substrate by each of the selected kinase (see screen-shot on the right; click to resize). Each bar-graph can be moved (mouse drag-and-drop).
The "contour only" bars are proportional with the String score (the blue scale at the top of the bar-graph correspond to scores for values of 0, 0.5 and 1). The number of distinct phosphorylation sites is displayed on the right as a number in square brackets [] followed by a bar-graph depicting the number of sites. An example is presented on the right (click to resize). Lastly, it is worth noting that, for each kinase, the predicted score (the horizontal length of the bar-graphs) is the probabilistic reunion of the scores for all phosphorylation sites. In our example MAPK8 has 3 phosphorylation sites and it's overall score is calculated as s=1-(1-s1)(1-s2)(1-s3) where s1,s2,s3 are the scores for each of the 3 phosphorylation sites.


submit Phospho Protein example - click to resize Submit Phospho Protein

The Submit Phospho Protein allows users to obtain Kinase predictions for proteins and phosphorylation sites of their choice.

The user must enter the protein sequences of interest in one field, in FASTA format. In a second field, the list of phosphorylation sites is expected and one can choose to include the known pELM sites or not. By default an example is always provided (click image to the left) when [Submit Phospho Protein] tab is selected; use [Clear Fields] button when needed validation of phopsho sites example - click to resize

Once the requested elements are introduced, the next stage is to validate them (see example on the right - click to enlarge). This is useful when mistakes are made (marked with question marks), for example when a wrong phosphorylation position is attributed to an amino acid other than S/T/Y.

After the validation, the web service starts the predictor and in a few moments the results are generated and displayed in a tabular form similar to the "Basic Search" result. Users can download the results in one of the "tab-separated-values" (.tsv) or Cytoscape (.sif) formats.


Institutes behind NetworKIN Institutes behind NetworKIN
(C) 2005-2010 Rune Linding and Lars Juhl Jensen
Web-development: Adrian Pasculescu and Marina Olhovsky

get Firefox browser Valid CSS! Valid XHTML 1.0 PostgreSQL Apache
Python.org