featureSelect {TSGS}

Trait specific gene selection using SVM and GA

Description

This function gives the optimal set of informative genes based on RNA-Seq count data

Usage

featureSelect(X, y, p = 5, n.iter = 1, alpha = 0.05, p.adj.method = "bonferroni")

Arguments

X

X is a G x N data frame of gene expression values (raw count data) where rows represent genes and columns represent samples. Each cell entry represents the read counts of of a gene in a sample (row names of X as gene names or gene ids)

y

y is a N x 1 numeric vector with entries 0 or 1 representing sample labels, where, 0/1 represents the sample label of samples for two conditions, e.g., 0 for Control and 1 for Case

p

Population size, by default 5

n.iter

The number of iterations, by default 1

alpha

The level of significance, by default 0.05

p.adj.method

Method of adjusting p-values, by default "bonferroni". The other methods available are "BH", "holm", "hochberg", "hommel", "BY".

Value

InformativeGenes

List of informative genes selected

LogCPM

Log cpm data of informative genes

DEA_Result

Differential Expression Analysis Result of informative genes

Author(s)

c(person("Md. Samir", "Farooqi", email = "[e-mail]", role = "aut"), person("K.K.", "Chaturvedi", email = "[e-mail]", role = "aut"), person("D.C.", "Mishra", email = "[e-mail]", role = "aut"), person("Sudhir", "Srivastava", email = "[e-mail]", role = c("cre","aut")))

Examples

  filename <- system.file("extdata", "exampleData.csv", package = "TSGS")
  cdata <- read.csv(filename, header = TRUE, row.names = 1, stringsAsFactors = FALSE)
  X <- as.data.frame(cdata[-1,])
  y <- as.numeric(cdata[1,])
  set.seed(100)
  result <- featureSelect(X, y, 5, 1, 0.05, "bonferroni")
  gene_list <- result$InformativeGenes
  logcpm_data <- result$LogCPM
  dea_result <- result$DEA_Result