GenerateTreemap: Generate Treemap

In dgrtwo/GSEAMA: Gene set enrichment analysis made awesome

Description

Visualize genes nested within specific and general GO sets using treemaps

Usage

GenerateTreemap(m, edges = NULL, ...)

Arguments

m	GeneMatrix object representing GO terms
...	Extra arguments passed on to threshold_sets

Details

Each gene should maximize the absolute sum of relevant node weights but every GO category will not be included since a gene can only be assigned to a single ancestry. A small GO category may fall into multiple ancestral paths some which are parents of other meaningful categories and others that are dead-ends. Should favor attachment to pathways with the most informative children. To allow this to occur, three edge weights are used 1) the path with the highest overall absolute score * number of genes affected is chosen 2) ancestors of LASSO nodes all get a smaller score: [min(abs(beta))]*0.01*N_lasso_children - applied to edge where ancestors are children 3) all remaining edges that are not predictive recieve a tiny weight so that they are not filtered: [min(abs(beta))]*1e-4

Value

Generates a treemap hierarchy Treemap GO Plot

Generate a treemap plot using the minimal spanning GO tree with genes assigned to individual GO terms.

Default color based on GO significance and effect sizes of individual genes

dgrtwo/GSEAMA documentation built on May 15, 2019, 7:22 a.m.