Removing points from geom_bar legend ggplot r - r

This is my data.
Mod <- as.factor(c(rep("GLM",5),rep("MLP",5),rep("RF",5),rep("SVML",5),rep("SVMR",5)))
Manifold <- as.factor(rep(c("LLE","Iso","PCA","MDS","kPCA"),5))
ROC <- runif(25,0,1)
Sens <- runif(25,0,1)
Spec <- runif(25,0,1)
df <- data.frame("Mod"= Mod, "Manifold"= Manifold, "ROC" = ROC, "Sens" = sens, "Spec" = spec)
And I am making this graph
resul3 <- ggplot(df, aes(x = Mod, y = ROC, fill= Manifold)) +
geom_bar(stat = "identity", position = "dodge", color = "black") +
ylab("ROC & Specificity") +
xlab("Classifiers") +
theme_bw() +
ggtitle("Classifiers' ROC per Feature Extraction Plasma") +
geom_point(aes(y=Spec), color = "black", position=position_dodge(.9)) +
scale_fill_manual(name = "Feature \nExtraction", values = c("#FFEFCA",
"#EDA16A" ,"#C83741", "#6C283D", "#62BF94"))
first graph
And what I want is another legend with tittle "Specificity" and a single black point. I dont want the point to be inside the Manifolds legend.
Something like this but without the points inside the manifold squares

Changing the geom_point line, adding a scale_color_manual and using the override as seen in #drmariod's answer will result in this plot:
ggplot(df, aes(x = Mod, y = ROC, fill= Manifold)) +
geom_bar(stat = "identity", position = "dodge", color = "black") +
ylab("ROC & Specificity") +
xlab("Classifiers") +
theme_bw() +
ggtitle("Classifiers' ROC per Feature Extraction Plasma") +
geom_point(aes(y=Spec, color = "Specificity"), position=position_dodge(.9)) +
scale_fill_manual(name = "Feature \nExtraction", values = c("#FFEFCA",
"#EDA16A" ,"#C83741", "#6C283D", "#62BF94")) +
scale_color_manual(name = NULL, values = c("Specificity" = "black")) +
guides(fill = guide_legend(override.aes = list(shape = NA)))

You can overwrite the aesthetics for shape and set it to NA like this
ggplot(df, aes(x = Mod, y = ROC, fill= Manifold)) +
geom_bar(stat = "identity", position = "dodge", color = "black") +
ylab("ROC & Specificity") +
xlab("Classifiers") +
theme_bw() +
ggtitle("Classifiers' ROC per Feature Extraction Plasma") +
geom_point(aes(y=Spec), color = "black", position=position_dodge(.9)) +
scale_fill_manual(name = "Feature \nExtraction", values = c("#FFEFCA",
"#EDA16A" ,"#C83741", "#6C283D", "#62BF94")) +
guides(fill = guide_legend(override.aes = list(shape = NA)))

Related

How to calculate standard error instead of standard deviation in ggplot

I need some help to figure out to estimate the standard error using the following R script:
library(ggplot2)
library(ggpubr)
library(Hmisc)
data("ToothGrowth")
ToothGrowth$dose <- as.factor(ToothGrowth$dose)
head(ToothGrowth, 4)
theme_set(
theme_classic() +
theme(legend.position = "top")
)
# Initiate a ggplot
e <- ggplot(ToothGrowth, aes(x = dose, y = len))
# Add mean points +/- SD
# Use geom = "pointrange" or geom = "crossbar"
e + geom_violin(trim = FALSE) +
stat_summary(
fun.data = "mean_sdl", fun.args = list(mult = 1),
geom = "pointrange", color = "black"
)
# Combine with box plot to add median and quartiles
# Change fill color by groups, remove legend
e + geom_violin(aes(fill = dose), trim = FALSE) +
geom_boxplot(width = 0.2)+
scale_fill_manual(values = c("#00AFBB", "#E7B800", "#FC4E07"))+
theme(legend.position = "none")
Many thanks for the help
Kind regards
A couple of things. First, you need to reassign e when you add geom_violin and stat_summary. Otherwise, it isn't carrying those changes forward when you add the boxplot in the next step. Second, when you add the boxplot last, it is mapping over the points and error bars from stat_summary so it looks like they're disappearing. If you add the boxplot first and then stat_summary the points and error bars will be placed on top of the boxplot. Here is an example:
library(ggplot2)
library(ggpubr)
library(Hmisc)
data("ToothGrowth")
ToothGrowth$dose <- as.factor(ToothGrowth$dose)
theme_set(
theme_classic() +
theme(legend.position = "top")
)
# Initiate a ggplot
e <- ggplot(ToothGrowth, aes(x = dose, y = len))
# Add violin plot
e <- e + geom_violin(trim = FALSE)
# Combine with box plot to add median and quartiles
# Change fill color by groups, remove legend
e <- e + geom_violin(aes(fill = dose), trim = FALSE) +
geom_boxplot(width = 0.2)+
scale_fill_manual(values = c("#00AFBB", "#E7B800", "#FC4E07"))+
theme(legend.position = "none")
# Add mean points +/- SE
# Use geom = "pointrange" or geom = "crossbar"
e +
stat_summary(
fun.data = "mean_se", fun.args = list(mult = 1),
geom = "pointrange", color = "black"
)
You said in a comment that you couldn't see any changes when you tried mean_se and mean_cl_normal. Perhaps the above solution will have solved the problem, but you should see a difference. Here is an example just comparing mean_se and mean_sdl. You should notice the error bars are smaller with mean_se.
ggplot(ToothGrowth, aes(x = dose, y = len)) +
stat_summary(
fun.data = "mean_sdl", fun.args = list(mult = 1),
geom = "pointrange", color = "black"
)
ggplot(ToothGrowth, aes(x = dose, y = len)) +
stat_summary(
fun.data = "mean_se", fun.args = list(mult = 1),
geom = "pointrange", color = "black"
)
Here is a simplified solution if you don't want to reassign at each step:
ggplot(ToothGrowth, aes(x = dose, y = len)) +
geom_violin(aes(fill = dose), trim = FALSE) +
geom_boxplot(width = 0.2) +
stat_summary(fun.data = "mean_se", fun.args = list(mult = 1),
geom = "pointrange", color = "black") +
scale_fill_manual(values = c("#00AFBB", "#E7B800", "#FC4E07")) +
theme(legend.position = "none")

Merge legend in ggplot when the geoms are different

I have the following code which yields the figure below:
ggplot(data=data.frame(x=x, y=y, mass=mass)) +
geom_line(mapping = aes(x=x, y=y, linetype='Gompertz predicted mass', col='Gompertz predicted mass')) +
geom_point(mapping = aes(x=x, y=mass, shape='Actual mass',col='Actual mass')) +
theme_bw() +
ylab('Mass') +
xlab('t') +
scale_color_manual(name='',values = c("black",'red')) +
scale_linetype_manual(name='',values = c("solid")) +
scale_shape_manual(name='', values = c(19)) +
scale_x_continuous(breaks=seq(4,26,2)) +
ylim(c(0, 20000)) +
ggtitle('Problem 3: Plot of tumor mass with time')
Notice how the legend is separated. I'd like to merge it for shape and color. When the geoms are the same, the technique of using scale_something_manual works perfectly fine to merge the legends. However, I'm having trouble with it here since I have two different geoms.
The problem is similar to the one described in https://github.com/tidyverse/ggplot2/issues/3648. There is no elegant solution at the moment. Because you haven't included any data, I've presumed that your problem is conceptually similar to the plot below:
library(ggplot2)
ggplot(mpg, aes(displ, hwy)) +
geom_point(aes(shape = "Point", colour = "Point")) +
geom_smooth(aes(linetype = "Line", colour = "Line"),
formula = y ~ x, se = FALSE, method = "loess") +
scale_colour_manual(values = c("red", "black")) +
scale_linetype_manual(values = "solid") +
scale_shape_manual(values = 19)
The way to fix the problem is to get rid of the linetype and shape aesthetics and scales, and instead override aesthetics at the level of the legend.
ggplot(mpg, aes(displ, hwy)) +
geom_point(aes(colour = "Point")) +
geom_smooth(aes(colour = "Line"),
formula = y ~ x, se = FALSE, method = "loess") +
scale_colour_manual(
values = c("red", "black"),
guide = guide_legend(override.aes = list(shape = c(NA, 19),
linetype = c(1, NA)))
)
Created on 2021-09-04 by the reprex package (v2.0.1)

Preventing wrong density plots when coloring histograms according to groups

based on some dummy data I created a histogram with desity plot
set.seed(1234)
wdata = data.frame(
sex = factor(rep(c("F", "M"), each=200)),
weight = c(rnorm(200, 55), rnorm(200, 58))
)
a <- ggplot(wdata, aes(x = weight))
a + geom_histogram(aes(y = ..density..,
# color = sex
),
colour="black",
fill="white",
position = "identity") +
geom_density(alpha = 0.2,
# aes(color = sex)
) +
scale_color_manual(values = c("#868686FF", "#EFC000FF"))
The histogram of weight shall be colored corresponding to sex, so I use aes(y = ..density.., color = sex) for geom_histogram():
a + geom_histogram(aes(y = ..density..,
color = sex
),
colour="black",
fill="white",
position = "identity") +
geom_density(alpha = 0.2,
# aes(color = sex)
) +
scale_color_manual(values = c("#868686FF", "#EFC000FF"))
As I want it to, the density plot stays the same (overall for both groups), but the histograms jump scale up (and seem to be treated individually now):
How do I prevent this from happening? I need individually colored histogram bars but a joint density plot for all coloring groups.
P.S.
Using aes(color = sex) for geom_density() gets everything back to original scales - but I don't want individual density plots (like below):
a + geom_histogram(aes(y = ..density..,
color = sex
),
colour="black",
fill="white",
position = "identity") +
geom_density(alpha = 0.2,
aes(color = sex)
) +
scale_color_manual(values = c("#868686FF", "#EFC000FF"))
EDIT:
As it has been suggested, dividing by the number of groups in geom_histogram()'s aesthetics with y = ..density../2 may approximate the solution. Nevertheless, this only works with symmetric distributions like in the first output below:
a + geom_histogram(aes(y = ..density../2,
color = sex
),
colour="black",
fill="white",
position = "identity") +
geom_density(alpha = 0.2,
) +
scale_color_manual(values = c("#868686FF", "#EFC000FF"))
which yields
Less symmetric distributions, however, may cause trouble using this approach. See those below, where for 5 groups, y = ..density../5 was used. First original, then manipulation (with position = "stack"):
Since the distribution is heavy on the left, dividing by 5 underestimates on the left and overestimates on the right.
EDIT 2: SOLUTION
As suggested by Andrew, the below (complete) code solves the problem:
library(ggplot2)
set.seed(1234)
wdata = data.frame(
sex = factor(rep(c("F", "M"), each = 200)),
weight = c(rnorm(200, 55), rnorm(200, 58))
)
binwidth <- 0.25
a <- ggplot(wdata,
aes(x = weight,
# Pass binwidth to aes() so it will be found in
# geom_histogram()'s aes() later
binwidth = binwidth))
# Basic plot w/o colouring according to 'sex'
a + geom_histogram(aes(y = ..density..),
binwidth = binwidth,
colour = "black",
fill = "white",
position = "stack") +
geom_density(alpha = 0.2) +
scale_color_manual(values = c("#868686FF", "#EFC000FF")) +
# Use fixed scale for sake of comparability
scale_x_continuous(limits = c(52, 61)) +
scale_y_continuous(limits = c(0, 0.25))
# Plot w/ colouring according to 'sex'
a + geom_histogram(aes(x = weight,
# binwidth will only be found if passed to
# ggplot()'s aes() (as above)
y = ..count.. / (sum(..count..) * binwidth),
color = sex),
binwidth = binwidth,
fill="white",
position = "stack") +
geom_density(alpha = 0.2) +
scale_color_manual(values = c("#868686FF", "#EFC000FF")) +
# Use fixed scale for sake of comparability
scale_x_continuous(limits = c(52, 61)) +
scale_y_continuous(limits = c(0, 0.25)) +
guides(color = FALSE)
Note:
binwidth = binwidth needed to be passed to ggplot()'s aes(), otherwise the pre-specified binwidth would not be found by geom_histogram()'s aes(). Further, position = "stack" is specified, so that both versions of the histogram are comparable. Plots for dummy data and the more complex distribution below:
Solved - Thanks for your help!
I don't think you can do it using y=..density.., but you can recreate the same thing like this...
binwidth <- 0.25 #easiest to set this manually so that you know what it is
a + geom_histogram(aes(y = ..count.. / (sum(..count..) * binwidth),
color = sex),
binwidth = binwidth,
fill="white",
position = "identity") +
geom_density(alpha = 0.2) +
scale_color_manual(values = c("#868686FF", "#EFC000FF"))

adding summary statistics to two factor boxplot

I would like to add summary statistics (e.g. mean) to the boxplot which have two factors. I have tried this:
library(ggplot2)
ggplot(ToothGrowth, aes(x = factor(dose), y = len)) +
stat_boxplot(geom = "errorbar", aes(col = supp, fill=supp), position = position_dodge(width = 0.85)) +
geom_boxplot(aes(col = supp, fill=supp), notch=T, notchwidth = 0.5, outlier.size=2, position = position_dodge(width = 0.85)) +
stat_summary(fun.y=mean, aes(supp,dose), geom="point", shape=20, size=7, color="violet", fill="violet") +
scale_color_manual(name = "SUPP", values = c("blue", "darkgreen")) +
scale_fill_manual(name = "SUPP", values = c("lightblue", "green"))
I got this picture:
It is possible somehow put the sample size of each box (e.g. top of the whiskers)? I have tried this:
ggplot(ToothGrowth, aes(x = factor(dose), y = len)) +
stat_boxplot(geom = "errorbar", aes(col = supp, fill=supp), position = position_dodge(width = 0.85)) +
geom_boxplot(aes(col = supp, fill=supp), notch=T, notchwidth = 0.5, outlier.size=2, position = position_dodge(width = 0.85)) +
stat_summary(fun.y=mean,aes(supp,dose),geom="point", shape=20, size=7, color="violet", fill="violet") +
scale_color_manual(name = "SUPP", values = c("blue", "darkgreen")) +
scale_fill_manual(name = "SUPP", values = c("lightblue", "green")) +
geom_text(data = ToothGrowth,
group_by(dose, supp),
summarize(Count = n(),
q3 = quantile(ToothGrowth, 0.75),
iqr = IQR(ToothGrowth),
aes(x= dose, y = len,label = paste0("n = ",Count, "\n")), position = position_dodge(width = 0.75)))
You can state the aesthetics just once by putting them in the main ggplot call and then they will apply to all of the geom layers: ggplot(ToothGrowth, aes(x = factor(dose), y = len, color=supp, fill=supp))
For the count of observations: The data summary step in geom_text isn't coded properly. Also, to set len (the y-value) for the text placement, the summarize function needs to output values for len.
To add the mean values in the correct locations on the x-axis, use stat_summary with the exact same aesthetics as the other geoms and stats. I've overridden the color aesthetic by setting the color to yellow so that the point markers will be visible on top of the box plot fill colors.
The code to implement the plot is below:
library(tidyverse)
pd = position_dodge(0.85)
ggplot(ToothGrowth, aes(x = factor(dose), y = len, color=supp, fill=supp)) +
stat_boxplot(geom = "errorbar", position = pd) +
geom_boxplot(notch=TRUE, notchwidth=0.5, outlier.size=2, position=pd) +
stat_summary(fun.y=mean, geom="point", shape=3, size=2, colour="yellow", stroke=1.5,
position=pd, show.legend=FALSE) +
scale_color_manual(name = "SUPP", values = c("blue", "darkgreen")) +
scale_fill_manual(name = "SUPP", values = c("lightblue", "green")) +
geom_text(data = ToothGrowth %>% group_by(dose, supp) %>%
summarize(Count = n(),
len=max(len) + 0.05 * diff(range(ToothGrowth$len))),
aes(label = paste0("n = ", Count)),
position = pd, size=3, show.legend = FALSE) +
theme_bw()
Note that the notch goes outside the hinges for all of the box plots. Also, having the sample size just above the maximum of each boxplot seems distracting and unnecessary to me. You could place all of the text annotations at the bottom of the plot like this:
geom_text(data = ToothGrowth %>% group_by(dose, supp) %>%
summarize(Count = n()) %>%
ungroup %>%
mutate(len=min(ToothGrowth$len) - 0.05 * diff(range(ToothGrowth$len))),
aes(label = paste0("n = ", Count)),
position = pd, size=3, show.legend = FALSE) +

changing the color palette in ggplot

I have created ggplot from my data (sample below):
I have created a violin plot of the NKV with the individual NKV data points plotted over it. I want to differentiate betweeen which PID my datapoints belong to. So far so good:
violin.murgang <- ggplot(nkv.murgang, aes(x = factor("Murgang"), nkv.murgang$NK)) +
geom_violin(color = "black", fill = "darkorange") +
ggtitle("NKV Murgang - Einfamilienhaus") +
labs(x = "Prozess", y = "Nutzen / Konsten \n Verhälhniss") +
stat_summary(geom = "text", fun.y = quantile,
aes(label=sprintf("%1.1f", ..y..)),
position=position_nudge(x=0.4), size=3) +
theme (legend.position = "none") +
stat_summary(fun.data = give.n, geom = "text", position=position_nudge(x=-0.4)) +
geom_jitter(aes(col = PID ), width = 0.35)
violin.murgang
The problem is that all the NKV data points are only visualized in different shade of blue. I would like to have different colours. I have tried adding this:
scale_colour_brewer(palette="Spectral")
which yields the error:
Error: Continuous value supplied to discrete scale
How can i achieve having different colour for the geom_jitter part?
What causes the error?
Thanks!
If you PID have more levels than colors of 'Spectral' palette, you could try scale_color_distiller, which extends brewer colors to continuous scale, see the manual of scale_color_distiller:
# Use distiller variant with continous data
v <- ggplot(faithfuld) +
geom_tile(aes(waiting, eruptions, fill = density))
v
v + scale_fill_distiller()
v + scale_fill_distiller(palette = "Spectral")
Therefore, we could try:
ggplot(nkv.murgang, aes(x = factor("Murgang"), nkv.murgang$NK)) +
geom_violin(color = "black", fill = "darkorange") +
ggtitle("NKV Murgang - Einfamilienhaus") +
labs(x = "Prozess", y = "Nutzen / Konsten \n Verhälhniss") +
stat_summary(geom = "text", fun.y = quantile,
aes(label=sprintf("%1.1f", ..y..)),
position=position_nudge(x=0.4), size=3) +
theme (legend.position = "none") +
geom_jitter(aes(color = PID), width = 0.35) +
scale_color_distiller(palette = "Spectral")
If you data has a few levels, we could use discrete scales. PID is integer, which does work with discrete scales. You should convert it to character or factor first:
ggplot(nkv.murgang, aes(x = factor("Murgang"), nkv.murgang$NK)) +
geom_violin(color = "black", fill = "darkorange") +
ggtitle("NKV Murgang - Einfamilienhaus") +
labs(x = "Prozess", y = "Nutzen / Konsten \n Verhälhniss") +
stat_summary(geom = "text", fun.y = quantile,
aes(label=sprintf("%1.1f", ..y..)),
position=position_nudge(x=0.4), size=3) +
theme (legend.position = "none") +
geom_jitter(aes(color = as.factor(PID) ), width = 0.35) +
scale_color_brewer(palette = "Spectral")

Resources