metsBoxplot: Creates layered boxplots of metabolite values across a class...

In buddha2490/BERGMets: Metabolomics data and analysis in the BERG

Description

Boxplots are often useful for illustrating the distribution of metabolite values across a categorical outcome variable. metsBoxplot() will create these boxplots using ggplot2. The plots are somewhat customizable, and can be further customizable by including a custom theme object.

Usage

metsBoxplot(dat, classvar, metabolite, ylog = F, boxcolors = NULL, ref_line = NULL, title, subtitle = NULL, xlab = NULL, ylab = NULL, xaxis_size = 12, yaxis_size = 12, titlesize = 14, plot_theme = NULL)

Arguments

dat	Data frame used for the analysis
classvar	Factor variable used for the x-axis
metabolite	Character vector (length==1) of the COMP_ID used for the y-axis. May be raw or log transformed
ylog	Logical, Plot y-axis on a log10 scale.
boxcolors	Vector of colors equal to the length of levels(classvar). If you leave this NULL, a black and white image will be returned, but you can color each of the boxes if needed.
ref_line	Numeric vector indicating a horizontal reference line. You may want a line to indicate overall minimum or median, or any other reference you might like to plot.
title	Character vector of a plot title, optional.
subtitle	Character vector of a subtitle, optional
xlab	X-axis label. Defaults to the classvar.
ylab	Y-axis label. Defaults to metabolite COMP_ID
xaxis_size	Font size for x-axis
yaxis_size	Font size for y-axis
titlesize	Font size for title
plot_theme	You can create your own ggplot2 theme object and include it here. If you leave it as NULL, the plot will return with Brian's preferred aesthetic tastes, for better or worse.

Value

ggplot2 object.

Author(s)

Brian Carter

Examples

breast <- getMetabolites("breast_metabolomics")

df <- left_join(survey,
                breast$metabolites,
                "ID")

df$BMICAT <- with(df,
                  ifelse(BMI < 18.5, 1, ifelse(
                       BMI < 25, 2, ifelse(
                            BMI < 30, 3, ifelse(
                                 BMI < 35, 4, 5)
                            )))) 
     factor(.,1:5, c("<18.5", "18.5-24.9",
                     "25-29.9", "30-34.9",
                     "35+"))

myplot <- metsBoxplot(dat=df,
            classvar="BMICAT",
            metabolite="X42459",
            ylog = T,
            boxcolors = c("royalblue","darkred","yellow",
                          "orange","forestgreen"),
            ref_line = min(df$X42459),
            title = "Title goes here",
            subtitle = "Subtitle goes here",
            xlab = "BMI Categories",
            ylab = "Sphingomyelin",
            xaxis_size = 12,
            yaxis_size = 12,
            titlesize = 14,
            plot_theme = NULL)


png("Boxplot.png", height=6, width=6, units="in", res=400)
myplot
dev.off()

buddha2490/BERGMets documentation built on Sept. 6, 2020, 5:11 p.m.