Drawing a boxplot for observed OTU count vs OTU picking method

Loading the libraries

library(reshape2)
library(ggplot2)

Loading the table onto R

data1 <- read.csv("summary_otu_count_for_boxplot.csv")

attach(data1)

The following objects are masked from data1 (pos = 3):

    gg.closed, gg.denovo, gg.open, silva.closed, silva.denovo, silva.open

The following objects are masked from data1 (pos = 4):

    gg.closed, gg.denovo, gg.open, silva.closed, silva.denovo, silva.open

names(data1)

[1] "silva.closed" "gg.closed"    "silva.denovo" "gg.denovo"    "silva.open"
[6] "gg.open"     

Here’s what the table contains:

data1

Generating a basic boxplot

ggplot(melt(data1), aes(variable, value)) + geom_boxplot()

No id variables; using all as measure variables

Adding some colour, axes labels and a legend

ggplot(melt(data1), aes(variable, value)) +
geom_boxplot(aes(color=factor(variable))) +
theme_bw() +
theme(legend.text = element_text(size = 9), legend.title = element_text(size = 9)) +
theme(axis.text = element_text(size = 9)) +
theme(axis.title = element_text(size = 9)) +
labs(x = "OTU picking method", y = "Observed OTU count",colour = "legend") +
annotation_logticks(sides = "rl") +
theme(panel.grid.minor = element_blank()) +
guides(title.hjust = 1.0) +
theme(plot.margin = unit(c(0,1,0,0),"mm"))

No id variables; using all as measure variables

Putting in a fill colour instead

ggplot(melt(data1), aes(variable, value)) +
  geom_boxplot(aes(fill = variable)) +
  theme_bw() +
  theme(legend.text = element_text(size = 9), legend.title = element_text(size = 9)) +
  theme(axis.text = element_text(size = 9)) +
  theme(axis.title = element_text(size = 9)) +
  labs(x = "OTU picking method", y = "Observed OTU count",colour = "legend") +
  annotation_logticks(sides = "rl") +
  theme(panel.grid.minor = element_blank()) +
  guides(title.hjust = 1.0) +
  theme(plot.margin = unit(c(0,1,0,0),"mm"))

No id variables; using all as measure variables