stocsy: Statistical TOtal Correlation SpectroscopY

In muma: Metabolomics Univariate and Multivariate Analysis

Description

Function to calculate correlation matrix and plot corresponding color-coded heatmap, useful for NMR molecular identification and assignment.

Usage

stocsy(scaling, threshold = FALSE, pos.threshold, neg.threshold)

Arguments

scaling	a character string indicating the name of the scaling previously used with the function 'explore.data'.
threshold	logical. The possibility to indicate or not a positive and negative threshold regulating the heatmap coloring.
pos.threshold	a decimal (or integer) indicating the upper level beyond which correlation values are visualized.
neg.threshold	a decimal (or integer) indicating the lower level below which correlation values are visualized.

Details

'threshold' is 'FALSE' by default. If 'TRUE' positive and negative threshold values must be specified. In this case two plots are created, one with no color thresholds and one with the indicated thresholds; both these heatmaps are written in the directory 'STOCSY', automatically created in the working directory.

Author(s)

Edoardo Gaude, Dimitrios Spiliotopoulos, Francesca Chignola, Silvia Mari, Andrea Spitaleri and Michela Ghitti

References

Cloarec, O et al. Statistical total correlation spectroscopy: an exploratory approach for latent biomarker identification from metabolic 1H NMR data sets. (2005) Anal Chem. 77(5):1282-9.

Examples

## The function is currently defined as
function (scaling, threshold = FALSE, pos.threshold, neg.threshold) 
{
    pwd.n = paste(getwd(), "/Preprocessing_Data_", scaling, "/ProcessedTable.csv", 
        sep = "")
    x <- read.csv(pwd.n, sep = ",", header = TRUE)
    x.x <- x[, 2:ncol(x)]
    rownames(x.x) <- x[, 1]
    x.t <- x.x
    mycor = cor(x.t, method = c("pearson"))
    library(gplots)
    col = colorpanel(50, "blue", "white", "red")
    image(mycor, axes = FALSE, col = col, main = "STOCSY")
    axis(side = 1, labels = colnames(mycor), at = seq(0, 1, length = length(colnames(mycor))), 
        las = 2, cex.axis = 0.7)
    axis(side = 2, labels = colnames(mycor), at = seq(0, 1, length = length(colnames(mycor))), 
        las = 2, cex.axis = 0.7)
    dirout = paste(getwd(), "/STOCSY/", sep = "")
    dir.create(dirout)
    o = paste(dirout, "STOCSY.pdf", sep = "")
    dev.copy2pdf(file = o)
    o.cor = paste(dirout, "CorrelationMatrix.csv", sep = "")
    write.csv(mycor, file = o.cor)
    if (threshold) {
        dev.new()
        image(mycor, axes = FALSE, zlim = c(pos.threshold, 1), 
            col = "red", main = paste("STOCSY <", neg.threshold, 
                " & >", pos.threshold, sep = ""))
        image(mycor, axes = FALSE, zlim = c(-1, neg.threshold), 
            col = "navy", add = TRUE)
        axis(side = 1, labels = colnames(mycor), at = seq(0, 
            1, length = length(colnames(mycor))), las = 2, cex.axis = 0.7)
        axis(side = 2, labels = colnames(mycor), at = seq(0, 
            1, length = length(colnames(mycor))), las = 2, cex.axis = 0.7)
        out = paste(dirout, "STOCSY_", pos.threshold, "_", neg.threshold, 
            ".pdf", sep = "")
        dev.copy2pdf(file = out)
    }
  }

muma documentation built on May 2, 2019, 9:45 a.m.