R/PCAPlot_2024.R

In eead-csic-eesa/fingerPro_model: Sediment Source Fingerprinting

#' Principal component analysis chart
#'
#' The function performs a principal components analysis on the given data matrix and displays a biplot using vqv.ggbiplot package of the results for each different source to help the user in the decision.
#'
#' @param data Data frame containing source and mixtures data
#' @param components Numeric vector containing the index of the two principal components in the chart
#' @param colors Vector of colors to use for the groups in the plot
#'
#' @export
#' 
PCAPlot <- function(data, components = c(1:2), colors = NULL) {
  # reorder groups
  data[, 2] <- factor(data[, 2], levels = unique(data[, 2]))
  data_PCA <- data[c(3:ncol(data))]
  ir.pca4 <- prcomp(data_PCA, center = TRUE, scale. = TRUE)
  
  p <- ggbiplot(ir.pca4, choices = components, obs.scale = 1, var.scale = 1, 
                ellipse = FALSE, groups = data[, 2], ellipse.prob = 0.95, circle = FALSE, 
                varname.size = 5, varname.adjust = 3, var.axes = TRUE) +
    geom_point(aes(shape = groups, colour = groups), size = 3) +
    geom_hline(yintercept = 0, colour = "black", linetype = "longdash", size = 1) +
    geom_vline(xintercept = 0, colour = "black", linetype = "longdash", size = 1) +
    theme(legend.direction = "horizontal", legend.position = "top") +
    stat_ellipse(type = "t", size = 1, geom = "polygon", alpha = 0.2, aes(fill = groups), level = 0.85) + 
    stat_ellipse(type = "t", size = 1, aes(colour = groups), level = 0.85) + 
    guides(groups = FALSE) +
    theme(text = element_text(size = 15)) + 
    ggtitle("PCA") +
    theme(plot.title = element_text(hjust = 0.5))
  
  # Apply custom colors if provided
  if (!is.null(colors)) {
    p <- p +
      scale_colour_manual(values = colors) +
      scale_fill_manual(values = colors)
  }
  
  return(p)
}