Doubts running ggplot 2: out-of-order data, and error bar

Question

Hello. I’m trying to use ggplot2 to make a graph on R. This graph is a bar graph that deals with various concentrations (each concentration has a bar) on the X axis, which is associated with a dependent variable on the Y axis which is parasitic growth inhibition (%).

However, I found a problem: It puts the concentrations out of order (The right one would be 13.07, then 6.53, then 3.26 etc).

The code I used

library(ggplot2)

comp_1_concentrations = c('13.07','6.53','3.26','1.63','0.81','0.4','0.2')

comp_1_inhibition = c(78,50,28,15,10,2,1.5)

comp_1_desv_pad = c(10,1,10,1.5,10,1,0.2)

data <- data.frame(
  name=comp_1_conc,
  value=comp_1_inhib,
  sd=comp_1_desv
)

ggplot(data) +
  geom_bar( aes(x=name, y=value), stat="identity", fill="black", alpha=1.7,color='black') + 
  geom_errorbar( aes(x=name, ymin=value, ymax=value+sd), width=0.5, colour="black", alpha=0.9, size=0.8)+theme_bw() + theme(panel.border = element_blank(), panel.grid.major = element_blank(),
                                                                                                                            panel.grid.minor = element_blank(), axis.line = element_line(colour = "black"))+labs(x = 'Concentrations (µM)')+labs(y = 'Growth inhibition (%)')+labs(title = 'Compound 1')

Answer 1

Let’s split the problems.

1) Change the order of the X-axis:

First of all, a small change of the data that has the variable name as a factor (that introduces some problems that will not help us now). stringsAsFactors = FALSE, keeps this variable as text. I imagine you want it like this.

data <- data.frame(
  name=comp_1_concentrations,
  value=comp_1_inhibition,
  sd=comp_1_desv_pad,
  stringsAsFactors = FALSE
)

With this new data, let’s go to the graph. The function reorder allows reordering the values of the first argument based on the values of the second argument. Turning them into numbers I believe they take the correct order.

p1 <- ggplot(data) +
  geom_bar(aes(x = reorder(name, as.numeric(name)), y = value), 
           stat = "identity", 
           fill = "black", 
           alpha = 0.7)
p1

2) Remove the background grid

To do this we use the function theme and their arguments. To better understand how they work, I recommend your documentation.

The argument panel.grid (with their respective divisions in major and minor and x and y) controls the background grids. To eliminate them we must assign element_blank() for him.

p2 <- p1 +
  theme(panel.grid = element_blank())
p2

3) Error bars and bars

In order for the error bars to show only the "positive" part of the error, simply use the value, and do not valor - sd as ymin. So we have:

p2 + 
  geom_errorbar(aes(x = name, ymin=value, ymax=value+sd), 
                width=0.5, colour="black", 
                alpha=0.9, size=1.3)