center

NAMESPACE

@@ -1,7 +1,7 @@
 # Generated by roxygen2: do not edit by hand
 
+export(center)
 export(preprocess)
-export(transform)
 import(dplyr)
 import(ggplot2)
 import(tidyr)

R/center.R

@@ -0,0 +1,50 @@
+#'
+#' Centering
+#'
+#' @description
+#' Apply centering to the data.
+#'
+#' @param dataSet A data frame containing the data signals.
+#'
+#' @param names A vector of strings (default = c("gender", "treatment", "replicate"))
+#' specifying the names of the attribute columns.
+#'
+#' @details
+#' The function executes the following:
+#' \enumerate{
+#' \item Plots the mean-variance relationship.
+#' \item Centers the data.
+#' \item Plots the mean-variance relationship again for comparison.
+#' }
+#'
+#' @returns The centered data.
+#'
+#' @autoglobal
+#'
+#' @export
+
+center <- function(dataSet,
+                   names = c("gender", "treatment", "replicate")) {
+
+  ## organize the data for centering
+  dataPoints <- dataSet %>%
+    select(-any_of(names))
+
+  ## calculate and plot a mean-variance plot
+  plotPre <- meanVarPlot(dataPoints, title = "Pre-Centering")
+  print(plotPre)
+
+  ## centering
+  centerData <- colMeans(dataPoints, na.rm=TRUE)
+  centeredDataPoints <- sweep(dataPoints, 2L, centerData, check.margin = FALSE)
+
+  ## calculate and plot a mean-variance plot
+  plotPost <- meanVarPlot(centeredDataPoints,  title = "Post-Centering")
+  print(plotPost)
+
+  ## recombine the labels and centered data into a single data frame
+  centeredDataSet <- cbind(dataSet[,names], centeredDataPoints)
+
+  ## return the centered data
+  return(centeredDataSet)
+}

R/globals.R

@@ -1,9 +1,9 @@
 # Generated by roxyglobals: do not edit by hand
 
 utils::globalVariables(c(
-  "Compound", # <preprocess>
-  "name", # <preprocess>
   "Mean", # <meanVarPlot>
   "Variance", # <meanVarPlot>
+  "Compound", # <preprocess>
+  "name", # <preprocess>
   NULL
 ))

R/meanVarPlot.R

@@ -0,0 +1,34 @@
+#'
+#' Plotting a graph of mean versus variance
+#'
+#' @description
+#' Take a set of metabolics data organized by column, calculate the mean and variance of
+#' each column, and then plot those statistics.
+#'
+#' @param datMV A data frame containing the data signals.
+#'
+#' @param title A string with the desired title for the mean-variance plot.
+#'
+#' @import ggplot2
+#' @importFrom stats var median
+#'
+#' @returns An object of class \code{plot}.
+#'
+#' @autoglobal
+#'
+#' @noRd
+
+meanVarPlot <- function(datMV, title = "") {
+
+  ## calculate the mean and variance for each protein individually
+  plotData <- data.frame(t(sapply(datMV, function(x) {
+    c(Mean = mean(x, na.rm = TRUE), Variance = var(x, na.rm = TRUE))
+  })))
+
+  ## plot the mean-variance relationship
+  ggplot(plotData, aes(Mean, Variance)) +
+    geom_point() +
+    labs(title = title) +
+    theme_bw() +
+    theme(plot.title = element_text(hjust = 0.5))
+}

docs/pkgdown.yml

@@ -2,7 +2,7 @@ pandoc: '3.2'
 pkgdown: 2.0.9
 pkgdown_sha: ~
 articles: {}
-last_built: 2024-06-02T03:07Z
+last_built: 2024-06-02T06:31Z
 urls:
   reference: https://uconn-scs.github.io/metastat/reference
   article: https://uconn-scs.github.io/metastat/articles