Reactome Pathway Summary Visualization Application

maulikkamdar edited this page Aug 27, 2012 · 4 revisions
Clone this wiki locally

The Reactome Pathway Summary Visualization Application pulls top-level pathway information from the Reactome Database using the Reactome REST API. It uses the frontPageItems API to get a list of all the top-level pathways and the associated child pathways (‘hasEvent’ element) and the pathwayParticipants API to get the total number of participating molecules in each pathway.

In all the layouts the top-level pathways are represented as circular nodes, and are linked to other top-level pathway nodes by edges, whose thickness is equal to the square-root of a similarity measure. This similarity measure is calculated by comparing the two pathways for the total number of common participating molecules between them.

Four different Kinds of Layout are currently provided for the user to visualize the Pathway Summary. Some features remain common between two layouts. The layout can easily be switched using the a select box provided in the user Interface. User Interface

A slider is also provided for the user to scroll and reduce the number of edges he wishes to see. Moving the slider sets a threshold, and all edges having a similarity measure less than that threshold are not visualized. The threshold can also be set using the text box provided.

Circular Layout :

Circular Layout This is the simplest layout and is the default layout of the application. Top-Level Pathway Nodes are arranged on the spokes of radii of a circle with fixed radius. The radius of the nodes is equal to the square root of the total number of participating molecules in the pathway. They are connected to other nodes by edges as seen from the figure. The thickness of the edge is the square root of the similarity measure.

Force-Directed Layout:

Force Directed Layout Initially the nodes are arranged in a circular layout and then are rearranged depending on the net force acting on each node. The layout is calculated over 100 iterations.

Each node has a force of Repulsion acting on it due to every other node which is calculated

1000 * (radius of node1 * radius of node2 )/ distance^2

and has a force of Attraction acting on it due to every edge connecting the node to other nodes which is calculated as

Edge_similarity * distance between the nodes/ 100

The net force is multiplied by a damping constant of 0.05 and the position of each node is adjusted accordingly. The video of the simulation can be seen here (HTML5 Video works in Chrome) or can be downloaded from here.

Hierarchical Layout:

Hierarchical Layout A Radial Treemap layout algorithm is implemented to show two layers of hierarchy. Top-level pathways are arranged in the first layer and first-level child pathways fetched from REST API are arranged in secondary layer. The nodes in a particular are of the same size, and this layout cannot be used to visualize total number of participating molecules in the top-level pathway.

Packed-Circle Layout (Experimental) :

Packed Circle Layout This layout is still experimental and subject to change. The idea is to arrange the child pathways in a circular layout inside the top-level pathway node, which here is of a fixed radius.

Expression Analysis

Expression Analysis can be carried out on the Reactome Pathway Summary Application workbench. The standard expression analysis form available on the website is just duplicated here and embedded in a window which would popup once the user clicks on the ‘Expression Analysis’ Button in the user interface. User Interface

The file upload function is still not available, however you can copy-paste the data-set in the text-box. The Popup Window is as shown below : Expression Popup Panel The remaining screen is completely disabled when the expression analysis is carried out, however a ‘Close’ button is provided on the Popup to quit the analysis workbench. Expression Analysis Once the expression is complete, the nodes are colored according to the percentage expression in the pathway. The values are mapped to a Blue-Yellow Color Scale and the user can select the time of the expression using radio-buttons that are now present on the interface.

The scale is also included in the user-interface to provide a clear idea to the user about the expression levels. The maximum and the minimum expression level is labelled on the scale. Pathways exhibiting negative expression levels would be colored in shades of blue, whereas pathways exhibiting positive expression levels would be colored in shades of yellow.