Mean per group on a bubble plot with ggplot - r

Question

I have a dataset with a lot of overlapping points and used ggplot to create a bubble plot to show that data. I need to add bars on my plot for the means of each group on the x axis (values can be 0, 1, or 2). I have tried to use geom_errorbar but haven't been able to get it to work with my data. Any help/suggestions would be greatly appreciated.
The following is my code and a script to generate fake data that is similar:
y <- seq(from=0, to=3.5, by=0.5)
x <- seq(from=0, to=2, by=1)
xnew <- sample(x, 100, replace=T)
ynew <- sample(y, 100, replace=T)
data <- data.frame(xnew,ynew)
data2 <- aggregate(data$xnew, by=list(x=data$xnew, y=data$ynew), length)
names(data2)[3] <- "Count"
ggplot(data2, aes(x = x, y = y)) +
geom_point(aes(size=Count)) +
labs(x = "Copies", y = "Score") +
aes(ymax=..y.., ymin=..y..) +
scale_x_continuous(breaks = seq(0, 2, 1)) +
scale_y_continuous(breaks = seq(0, 3, 0.5)) +
theme(legend.position = "bottom", legend.direction = "horizontal",
axis.line = element_line(size=1, colour = "black"),
panel.grid.major = element_blank(),
panel.grid.minor = element_blank(),
panel.border = element_blank(),
panel.background = element_blank(),
axis.text.x = element_text(colour="black", size = 10),
axis.text.y = element_text(colour="black", size = 10))

Answer 1

I am not entirely sure that I understand your question correctly. It seems to me that in addition to the bubbles, you want to visualise the mean value of y for each value of x as a bar of some kind. (You mention error bars, but it seems that this is not a requirement, but just what you have tried. I will use geom_col() instead.)
I assume that you want to weigh the mean over y by the counts, i.e., sum(y * Count) / sum(Count). You can create a data frame that contains these values by using dplyr:
data2_mean
## # A tibble: 3 × 2
## x y
## <dbl> <dbl>
## 1 0 1.833333
## 2 1 1.750000
## 3 2 2.200000
When creating the plot, I use data2 as the data set for geom_point() and data2_mean as the data set for geom_col(). It is important to put the bars first, since the bubbles should be on top of the bars.
ggplot() +
geom_col(aes(x = x, y = y), data2_mean, fill = "gray60", width = 0.7) +
geom_point(aes(x = x, y = y, size = Count), data2) +
labs(x = "Copies", y = "Score") +
scale_x_continuous(breaks = seq(0, 2, 1)) +
scale_y_continuous(breaks = seq(0, 3, 0.5)) +
theme(legend.position = "bottom", legend.direction = "horizontal",
axis.line = element_line(size=1, colour = "black"),
panel.grid.major = element_blank(),
panel.grid.minor = element_blank(),
panel.border = element_blank(),
panel.background = element_blank(),
axis.text.x = element_text(colour="black", size = 10),
axis.text.y = element_text(colour="black", size = 10))
Everything that I changed compared to your code comes before scale_x_continuous(). This produces the following plot:

Answer 2

Is this what you're after? I first calculated the group-level means using the dplyr package and then added line segments to your plot using geom_segment:
library(ggplot2)
library(dplyr)
data2 <- data2 %>% group_by(x) %>% mutate(mean.y = mean(y))
ggplot(data2, aes(x = x, y = y)) +
geom_point(aes(size=Count)) +
labs(x = "Copies", y = "Score") +
aes(ymax=..y.., ymin=..y..) +
scale_x_continuous(breaks = seq(0, 2, 1)) +
scale_y_continuous(breaks = seq(0, 3, 0.5)) +
theme(legend.position = "bottom", legend.direction = "horizontal",
axis.line = element_line(size=1, colour = "black"),
panel.grid.major = element_blank(),
panel.grid.minor = element_blank(),
panel.border = element_blank(),
panel.background = element_blank(),
axis.text.x = element_text(colour="black", size = 10),
axis.text.y = element_text(colour="black", size = 10)) +
geom_segment(aes(y = mean.y, yend = mean.y, x = x -0.25, xend = x + 0.25))