| title | author | date | vignette | output | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
amarettoTools -- working with AMARETTO outputs |
Vincent J. Carey, stvjc at channing.harvard.edu |
`r format(Sys.time(), '%B %d, %Y')` |
%\VignetteEngine{knitr::rmarkdown} %\VignetteIndexEntry{amarettoTools -- working with AMARETTO outputs} %\VignetteEncoding{UTF-8}
|
|
Introduction
library(BiocStyle)
library(amarettoTools)
AMARETTO is an integrative algorithm for identifying driver genes in multiomic cancer studies. In this package we address the structure of AMARETTO results.
An illustration in glioblastoma
The main result is a collection of modules of coexpressed and coregulated genes. The structure of the collection can be captured in a graphNEL instance as defined in the Bioconductor graph package.
library(amarettoTools)
data(regulGBM)
regulGBM
# how many modules
length(grep("Module", nodes(regulGBM)))
# target genes in Module_1
adj(regulGBM, "Module_1")
# regulators of Module_1
inEdges("Module_1", regulGBM)
# weights for regulators of module 1
m1regs = inEdges("Module_1", regulGBM)[[1]]
t(t(sapply(edgeWeights(regulGBM, m1regs), "[", "Module_1")))
Conclusions
This graph losslessly collects the relationship information in the AMARETTO run. The expression data are propagated but as far as I can tell, they are just the expression quantifications from firehose? So they should not be propagated but queried from the original source.