R ggplot multi-row x-axis labels

Question

This is my data (in dput format):

    structure(list(Biological.Group = structure(1:15, .Label = c("A", 
"C", "E", "D", "F", "G", "W", "R", "T", "Y", "V", "B", "N", "M", 
"L"), class = "factor"), Group = c(NA, NA, 1L, 1L, 1L, 1L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L), Relative.Normalized.Expression....Cq. = c(5L, 
7L, 3L, 5L, 6L, 8L, 9L, 1L, 4L, 5L, 7L, 4L, 2L, 7L, 8L), SE = c(0.171499719, 
0.089692493, 0.153777208, 0.188012958, 0.153776128, 0.192917199, 
0.224766056, 0.231338325, 0.121716906, 0.094763028, 0.09635363, 
0.069986333, 0.113681329, 0.094614957, 0.391182473)), row.names = c(NA, 
-15L), class = "data.frame")

This is the code for my graph:

library(ggplot2)
colours <- c("A"="darkolivegreen4","C"="darkolivegreen2","E"="darkseagreen4","D"="darkseagreen3", "F"="gold3","G"="gold", "W"="goldenrod2","R"="yellow2","T"="mediumpurple","Y"="thistle1", "V"="turquoise","B"="pink3", "N"="plum4", "M"="tomato3", "L"="orange1")
df$Biological.Group <- as.character(df$Biological.Group)
df$Biological.Group <- factor(df$Biological.Group, 
levels=unique(df$Biological.Group))
plot = ggplot(df, aes(Biological.Group,Relative.Normalized.Expression....Cq.,fill=factor(Biological.Group)))+ scale_fill_manual(values = colours) + xlab("Condition") +ylab("Relative Normalized Expression (∆∆Cq)")+geom_bar(stat="identity")
plot + geom_errorbar(aes(ymin=Relative.Normalized.Expression....Cq.-SE,ymax=Relative.Normalized.Expression....Cq.+SE),width=0.5)+geom_boxplot() + theme(axis.text.x = element_text(angle = 60, hjust = 1)) + theme(legend.position="none") +ggtitle("Mrdf-1")+theme(plot.title = element_text(size=12, face="italic",hjust=0.5))
ggsave("dfGraph.png",height=5.64,width=5.64)

This is what the graph looks like:

I want to include the "Group" column data (2nd column of data) as secondary labels in the x axis like this:

If possible, I'd also like to make the space between groups bigger (meaning between "A" and "C", "C" and "E", and "G" and "W")

Answer 1

I've made some alterations to your ggplot().

• there is no need to convert using factor()
• there is no need to use geom_boxplot()
• geom_col() is less typing than geom_bar(stat = "identity")
• no need for multiple theme() lines
• I used labs() for x, y and title labels

The labeling of Group can be achieved using facet_grid().

Not sure what you mean by "space between groups" - from your question it sounds more like "space between bars" - which can be achieved by adjusting width.

My code:

library(dplyr)

df %>% 
  mutate(Group = ifelse(is.na(Group), "", Group)) %>% 
  ggplot(aes(Biological.Group, 
             Relative.Normalized.Expression....Cq., 
             fill = Biological.Group)) +
  geom_col(width = 0.7) + 
  geom_errorbar(aes(ymin = Relative.Normalized.Expression....Cq.-SE,
                    ymax = Relative.Normalized.Expression....Cq.+SE), width = 0.5) +
  scale_fill_manual(values = colours) + 
  labs(x = "Condition",
       y = "Relative Normalized Expression (∆∆Cq)",
       title = "Mrdf-1") +
  theme(axis.text.x = element_text(angle = 60, 
                                   hjust = 1),
        legend.position = "none",
        plot.title = element_text(size = 12, 
                                  face = "italic", 
                                  hjust = 0.5),
        strip.background.x = element_blank()) + 
  facet_grid(~Group, 
             scales = "free_x", 
             switch = "x", 
             space = "free_x")

Result: