functions.R

# Function for creating a common legend for 2 ggplot2 figures.
grid_arrange_shared_legend <- function(..., ncol = length(list(...)), nrow = 1, position = c("bottom", "right")) {
  
  plots <- list(...)
  position <- match.arg(position)
  g <- ggplotGrob(plots[[1]] + theme(legend.position = position))$grobs
  legend <- g[[which(sapply(g, function(x) x$name) == "guide-box")]]
  lheight <- sum(legend$height)
  lwidth <- sum(legend$width)
  gl <- lapply(plots, function(x) x + theme(legend.position="none"))
  gl <- c(gl, ncol = ncol, nrow = nrow)
  
  combined <- switch(position,
                     "bottom" = arrangeGrob(do.call(arrangeGrob, gl),
                                            legend,
                                            ncol = 1,
                                            heights = unit.c(unit(1, "npc") - lheight, lheight)),
                     "right" = arrangeGrob(do.call(arrangeGrob, gl),
                                           legend,
                                           ncol = 2,
                                           widths = unit.c(unit(1, "npc") - lwidth, lwidth)))
  grid.newpage()
  grid.draw(combined)
  
}

### Functions for extracting the predicted R squared from lm models.
model_fit_stats <- function(linear.model) {
  r.sqr <- summary(linear.model)$r.squared
  adj.r.sqr <- summary(linear.model)$adj.r.squared
  pre.r.sqr <- pred_r_squared(linear.model)
  PRESS <- PRESS(linear.model)
  return.df <- data.frame(r.squared = r.sqr, adj.r.squared = adj.r.sqr, pred.r.squared = pre.r.sqr, press = PRESS)
  return(return.df)
}

pred_r_squared <- function(linear.model) {
  #' Use anova() to get the sum of squares for the linear model
  lm.anova <- anova(linear.model)
  #' Calculate the total sum of squares
  tss <- sum(lm.anova$'Sum Sq')
  # Calculate the predictive R^2
  pred.r.squared <- 1-PRESS(linear.model)/(tss)
  
  return(pred.r.squared)
}

PRESS <- function(linear.model) {
  #' calculate the predictive residuals
  pr <- residuals(linear.model)/(1-lm.influence(linear.model)$hat)
  #' calculate the PRESS
  PRESS <- sum(pr^2)
  
  return(PRESS)
}

##### Normalization #######

# Better rounding function than R's base round
myround <- function(x) { trunc(x + 0.5) }


# Scales reads by 
# 1) taking proportions
# 2) multiplying by a given library size of n
# 3) rounding 
# Default for n is the minimum sample size in your library
# Default for round is floor
scale_reads <- function(physeq, n = min(sample_sums(physeq)), round = "floor") {
  
  # transform counts to n
  physeq.scale <- transform_sample_counts(physeq, 
                                          function(x) {(n * x/sum(x))}
  )
  
  # Pick the rounding functions
  if (round == "floor"){
    otu_table(physeq.scale) <- floor(otu_table(physeq.scale))
  } else if (round == "round"){
    otu_table(physeq.scale) <- myround(otu_table(physeq.scale))
  }
  
  # Prune taxa and return new phyloseq object
  physeq.scale <- prune_taxa(taxa_sums(physeq.scale) > 0, physeq.scale)
  return(physeq.scale)
}

# Extract Legend from ggplot2 object
g_legend <- function(a.gplot){ 
  tmp <- ggplot_gtable(ggplot_build(a.gplot)) 
  leg <- which(sapply(tmp$grobs, function(x) x$name) == "guide-box") 
  legend <- tmp$grobs[[leg]] 
  return(legend)
  }