title: "Introduction to KEGGlincs" author: "Shana White" date: "03/23/2017"

by Shana White


#Make sure that the following bioconductor packages are installed
biocLite(c("hgu133a.db", "KEGGgraph", "KEGGREST", "KOdata"))

#Download package

#Load/activate package for use

Overview of package

KEGGlincs is an R package that provides a seamless interface for viewing detailed versions of KEGG pathways in Cytoscape such that the exact relationships (edges) that exist between pathway elements (nodes) are visualized in a meaningful manner.

This package is intended to be used as a tool to accurately regenerate KEGG pathways with expanded edges that are marked-up (in terms of color or color+width) to portray specific edge attributes. Specifically, - the edges are expanded to represent all relationships between genes encoded by the KGML file but not necessarily represented in the original [KEGG] pathway. This 'expansion' occurs either because: - one node may represent many [often paralogous] genes. - a node is of type 'group' and represents two or more genes that are part of the same functional complex. - the edge attributes represented can be specified by the user as either: - Functional - the exact type and [in some cases] subtype of the relationship defined by the KGML file. - Data driven - based on quantitative experimental evidence for gene-gene relationships generated by experimental data. As an added feature and motivating example, this package gives users an opportunity to use LINCS L1000 knock-out data in order to analyze/visualize the pathway edges as they pertain to specific cell lines.

Aside from edge-specific annotation/mark-up, a unique feature of this package lies in the ability for users to generate graph objects in R and send them to Cytoscape for an interactive visualization experience via utilities developed by the cyREST team.

The following example showcases the ideas described above.

Example: p53 pathway

Original KEGG pathway:

p53 pathway from KEGG website

Using KEGGlincs without adding data:

Detailed pathway visualized in Cytoscape

Note: An open Cytoscape session is required for pathway visualization; i.e. make sure that Cytoscape is open before running the following commands


Overlay pathway edges with LINCS data:

Pathway with edge attributes generated from LINCS L1000 knock-out data, refined by cell-line specific expression (default option, see below), and visualized in Cytoscape

Pathway specific to PC3 cell-line (Prostate cancer)

KEGG_lincs("hsa04115", "PC3")

Pathway specific to HA1E cell-line (Immortalized normal kidney epithelial)

KEGG_lincs("hsa04115", "HA1E")

uc-bd2k/KEGGlincs documentation built on Dec. 8, 2017, 3:32 a.m.