0_Dendogram.Rmd

# 1. Load libraries

```{r}
library(tidyverse)
library(ggdendro)
```


# 2. Dendogram


```{r}
# Import data

# Feature_correspondence
features <- read.csv("../featureCorrespondence_MS2.csv", row.names = 1)
# transpose
features <- t(features)
#features <- features[-c(7,8),]

# Metadata
metadata <- read.csv("../metadata.csv")
#metadata <- metadata[-c(7,8),]

# Normalize data, row wise normalisation
#dat.n <- scale(t(features))
dat.n <- t(features)
# Put data back in original form
dat.tn <- t(dat.n)

# Calculate distance between experiments in rows
d1 <- dist(dat.n, method = "euclidean", diag = FALSE, upper = FALSE)
d2 <- dist(dat.tn, method = "euclidean", diag = FALSE, upper = TRUE)

# Clustering distance between experiments using Ward linkage
c1 <- hclust(d1, method = "ward.D2")
# Clustering distance between proteins using Ward linkage
c2 <- hclust(d2, method = "ward.D2")

# Build dendogram object from hclust results
dend_c2 <- as.dendrogram(c2)
# Extract the data (for rectangular lines)
# Type can be "rectangle" or "triangle"
dend_data <- dendro_data(dend_c2, type = "rectangle")
# What contains dend_data
names(dend_data)
#head(dend_data$labels)
dend_labels <- dend_data$labels
dend_labels$label <- gsub(".", '-', dend_labels$label, fixed = T)
dend_data$labels <- merge(dend_labels, metadata, by.x ='label', by.y = 'SampleCode')

listColors <- c('Brevideciduous' = '#AA3377', 'Deciduous' = '#117733')
listShapes <- c('Red' = 19, 'White' = 17)

# Plot line segments and add labels
p <- ggplot() + 
  geom_segment(data=dend_data$segments, aes(x = x, y = y, xend = xend, yend = yend)) +
  geom_point(data=dend_data$labels, aes(x=x, y=y-3, shape=OakType, color=LeafLife), size=4) + 
  #geom_point(data=dend_data$labels, aes(x=x, y=-3, shape=trt), size=2) + 
  geom_text(data = dend_data$labels, aes(x, y=-6, label = itSpeciesName), parse = T, 
            hjust=0, size = 5, color='black', fontface="bold", show.legend = F) +
  coord_flip()  +
  scale_y_reverse(expand = c(0, 70)) +
  scale_x_continuous(breaks = dend_data$labels$x, 
                     labels = dend_data$labels$SpeciesName, position = "top") +
  scale_colour_manual(values=listColors) +
  scale_shape_manual(values = listShapes) + 
  theme_minimal(base_size = 10) +
  theme(axis.title = element_blank(), 
        # axis.text.x = element_text(), 
        axis.ticks.x = element_line(), 
        axis.line.x = element_line(),
        axis.text.y = element_text( face="bold", color="white"), 
        panel.grid = element_blank(),  
        legend.title = element_blank(),
        legend.position = "top") 
print(p)
ggsave('Dendogram_Features_euclidean_hclust_SpeciesName.png',units=c('in'),width=9,height=6,dpi=300,p)


p <- ggplot() + 
  geom_segment(data=dend_data$segments, aes(x = x, y = y, xend = xend, yend = yend)) +
  geom_point(data=dend_data$labels, aes(x=x, y=y-3, shape=OakType, color=LeafLife), size=3) + 
  #geom_point(data=dend_data$labels, aes(x=x, y=-3, shape=trt), size=2) + 
  geom_text(data = dend_data$labels, aes(x, y=-6, label = label, color=OakType), 
            hjust=0, size = 3, color = 'black', fontface="bold", show.legend = F) +
  coord_flip()  +
  scale_y_reverse(expand = c(0, 30)) +
  scale_x_continuous(breaks = dend_data$labels$x, 
                     labels = dend_data$labels$SpeciesName, position = "top") +
  scale_colour_manual(values=listColors) +
  scale_shape_discrete(name="Oak Type") + 
  theme_minimal() +
  theme(axis.title = element_blank(), 
        axis.text.x = element_text(size=10), 
        axis.ticks.x = element_line(), 
        axis.line.x = element_line(),
        axis.text.y = element_text(size = 10, face="bold", color="white"), 
        panel.grid = element_blank(),  
        legend.title = element_blank(),
        legend.position = "top") 
print(p)
ggsave('Dendogram_Features_euclidean_hclust_SampleCode.png',units=c('in'),width=9,height=6,dpi=300,p)

```