matchGeneSets: Creates a grouping of variables (genes) from gene sets

In markvdwiel/GRridge: Better prediction by use of co-data: Adaptive group-regularized ridge regression

Description

Creates a grouping of variables (genes) from gene sets by matching the IDs of the genes with the IDs of the members of the gene sets

Usage

matchGeneSets(GeneIds, GeneSets, minlen = 25, remain = TRUE)

Arguments

GeneIds	Character vector. Vector of gene IDs. Can be any ID (gene symbol, entrezID, etc) as long as it matches with IDs used in GeneSets.
GeneSets	Named list of character vectors. Each component of the list represents a named gene set. Each vector a list of member genes.
minlen	Integer. Minimum number of members of a gene set.
remain	Boolean. If remain=TRUE, all genes that are not in any list will be grouped in one remainder group.

Details

About minlen: to avoid overfitting in the grridge function, we recommend to not use groups with less than 25 members, unless monotone=TRUE is used in the grridge function, in which case 10 members may suffice as a lower bound. About remain: it is often beneficial to down-weight genes that are not part of any gene set, so we recommend to use remain=TRUE

Value

A list the components of which contain the indices of the variables belonging to each of the groups.

Author(s)

Mark A. van de Wiel

See Also

grridge, CreatePartition

Examples

# Load data objects
data(dataWurdinger)

# Transform the data set to the square root scale
dataSqrtWurdinger <- sqrt(datWurdinger_BC)

#Standardize the transformed data
datStdWurdinger <- t(apply(dataSqrtWurdinger,1,function(x){(x-mean(x))/sd(x)}))

# A list of gene names in the primary RNAseq data
genesWurdinger <- as.character(annotationWurdinger$geneSymbol)

# We  show an example of GRridge classification model by using overlapping groups,
# i.e. pathway-based grouping. Transcription factor based pathway was extracted from 
# the MSigDB (Section C3: motif gene sets; subsection: transcription factor targets;
# file's name: "c3.tft.v5.0.symbols.gmt"). The gene sets are based on 
# TRANSFAC version 7.5 database (http://www.gene-regulation.com/). 

# Some features may belong to more than one group. The argument minlen=25 implies 
# the minimum number of features in a gene set. If remain=TRUE, gene sets with less 
# than 25 members are grouped to the "remainder" group. "genesWurdinger" is an object
# containing gene names from the mRNA sequencing data set. 
# See help(matchGeneSets) for more detail information.
# Also see Vignette for more detail examples

## The "TFsym" is available on https://github.com/markvdwiel/GRridgeCodata
# gseTF <- matchGeneSets(genesWurdinger,TFsym,minlen=25,remain=TRUE)

markvdwiel/GRridge documentation built on May 21, 2019, 12:25 p.m.