sweep

knitr::opts_chunk$set(echo = TRUE)

https://bioinfomagician.wordpress.com/2014/08/12/my-favorite-commands-part3-sweep-function-in-r/

sweep()

• apply different values to data in different col/rows

data <- matrix(seq(1,12),ncol=4,nrow=3,byrow=TRUE)
data
sweep(data, 2, c(3, 4, 5, 6), "-")

example:

• standardize gene expression using sweep
• (normalized expression - median expression)/median absolute deviation
  • similar to zscore standardization ((x - mean)/sd)
• need to subtract different median expression (different for each gene),
  • and divide different MAD (different for each gene),
  • sweep() can help

standardize <- function(z) {
  rowmed <- apply(z, 1, median)
  rowmad <- apply(z, 1, mad)  # median absolute deviation
  rv <- sweep(z, 1, rowmed,"-")  #subtracting median expression
  rv <- sweep(rv, 1, rowmad, "/")  # dividing by median absolute deviation
  return(rv)
}
data <- data.frame(data)
colnames(data) <- c("sample1","sample2","sample3","sample4")
rownames(data) <- c("gene1","gene2","gene3")

data[1,] %>% class()
data[1,] %>% as.matrix() %>% median()
data[1,] %>% as.matrix() %>% mad()
((data[1,1]) - (data[1,] %>% as.matrix() %>% median()))/data[1,] %>% as.matrix() %>% mad()

print(data)
print(standardize(data))

EOF