tutorial_script.R

# GOAL: To visualise plant trait correlations and make a multi-plot panel of summary statistics

# Raw data: Measurements of several plant functional traits 

# (SLA, Leaf area, Leaf fresh mass, and Leaf dry mass) 
# on multiple individuals of four different species

# Get data
load("traits.RData")
load("traits_sum.RData")

# Load libraries
library(corrplot)
library(ggplot2)

# Comparing within-individual correlations
# Trait-trait correlation plot
(correlation <- corrplot(cor(traits[,2:5], use = "pairwise.complete.obs")))

# Save the plot in your working directory
png(filename = "trait_correlation.png", width = 600, height = 600)
(correlation <- corrplot(cor(traits[,2:5], use = "pairwise.complete.obs")))
dev.off()

# Transform dataset to long format
traits_long <- gather(traits, Trait, value, select = 2:5)

# Graph raw trait data behind mean +/- 95% CI's and save the file
(trait.plot <- ggplot()+
    geom_point(data = traits_long, mapping = aes(x = SpeciesName, y = value, colour = Trait), alpha = 0.1) +
    geom_errorbar(data = traits_sum, mapping = aes(x = SpeciesName, ymin = q2.5, ymax = q97.5, group = Trait), width = 0.3) +
    geom_point(data = traits_sum, mapping = aes(x = SpeciesName, y = mean, group = Trait), size = 4, colour = "black") +
    geom_point(data = traits_sum, mapping = aes(x = SpeciesName, y = mean, colour = Trait), size = 3) +
    facet_wrap(~Trait, scales = "free_y")+
    theme_classic() +
    scale_x_discrete(labels = c("Dryas", "Eriophorum", "Oxyria", "Salix")) +
    ylab("Trait Value") +
    xlab("Species"))

# We can save plots made using ggplot2 with ggsave, which is just one line of code
ggsave(trait.plot, filename = "traits.png", height = 5, width = 10)