buildNetworkPlot: Build a network plot of gene sets

In EvgeniyaGorobets/PGxVision: Visualize Biomarkers and Pharmacogenomic Data

Description

Create a network plot where each node corresponds to a gene set and edge weights and lengths correspond to the similarity score between two gene sets. The purpose is to visualize which gene sets share the most genes. Edges with low similarity scores can be pruned to help naturally divide gene sets into groups.

Usage

buildNetworkPlot(gsSimilarityDf, similarityCutoff = 0.5, title = NULL)

Arguments

gsSimilarityDf	A non-empty data.frame listing the similarity score for each pair of gene sets (essentially a table of edges). Must have columns "gs1", "gs2", "similarity". It is ok to provide edges between a node and itself, e.g., in the case of network plots with a single node. Such edges will be removed.
similarityCutoff	(optional) A number between 0 and 1 indicating the minimum similarity two gene sets must have in order for an edge to show up on the plot. Defaults to 0.5. Very low values will cause a highly connected, possibly confusing plot. Very high values may remove or mask natural gene set groupings.
title	(optional) A custom title for the network plot. Defaults to "Gene Set Similarity Plot"

Value

An interactive visNetwork object containing the network plot of gene sets.

Examples

result <- geneSetAnalysis("ENSG00000000971", "GO:BP")
buildNetworkPlot(result$similarityDf, similarityCutoff = 0.3)

EvgeniyaGorobets/PGxVision documentation built on Dec. 17, 2021, 7:26 p.m.