Related
I'd like to draw bar plot like this but in dual Y axis
(https://i.stack.imgur.com/ldMx0.jpg)
the first three indexs range from 0 to 1,
so I want the left y-axis (corresponding to NSE, KGE, VE) to range from 0 to 1,
and the right y-axis (corresponding to PBIAS) to range from -15 to 5.
the following is my data and code:
library("ggplot2")
## data
data <- data.frame(
value=c(0.82,0.87,0.65,-3.39,0.75,0.82,0.63,1.14,0.85,0.87,0.67,-7.03),
sd=c(0.003,0.047,0.006,4.8,0.003,0.028,0.006,4.77,0.004,0.057,0.014,4.85),
index=c("NSE","KGE","VE","PBIAS","NSE","KGE","VE","PBIAS","NSE","KGE","VE","PBIAS"),
period=c("all","all","all","all","calibration","calibration","calibration","calibration","validation","validation","validation","validation")
)
## fix index sequence
data$index <- factor(data$index, levels = c('NSE','KGE','VE',"PBIAS"))
data$period <- factor(data$period, levels = c('all','calibration', 'validation'))
## bar plot
ggplot(data, aes(x=index, y=value, fill=period))+
geom_bar(position="dodge", stat="identity")+
geom_errorbar(aes(ymin=value-sd, ymax=value+sd),
position = position_dodge(0.9), width=0.2 ,alpha=0.5, size=1)+
theme_bw()
I try to scale and shift the second y-axis,
but PBIAS bar plot was removed because of out of scale limit as follow:
(https://i.stack.imgur.com/n6Jfm.jpg)
the following is my code with dual y axis:
## bar plot (scale and shift the second y-axis with slope/intercept in 20/-15)
ggplot(data, aes(x=index, y=value, fill=period))+
geom_bar(position="dodge", stat="identity")+
geom_errorbar(aes(ymin=value-sd, ymax=value+sd),
position = position_dodge(0.9), width=0.2 ,alpha=0.5, size=1)+
theme_bw()+
scale_y_continuous(limits = c(0,1), name = "value", sec.axis = sec_axis(~ 20*.- 15, name="value"))
Any advice for move bar_plot or other solution？
Taking a different approach, instead of using a dual axis one option would be to make two separate plots and glue them together using patchwork. IMHO that is much easier than fiddling around with the rescaling the data (that's the step you missed, i.e. if you want to have a secondary axis you also have to rescale the data) and makes it clearer that the indices are measured on a different scale:
library(ggplot2)
library(patchwork)
data$facet <- data$index %in% "PBIAS"
plot_fun <- function(.data) {
ggplot(.data, aes(x = index, y = value, fill = period)) +
geom_bar(position = "dodge", stat = "identity") +
geom_errorbar(aes(ymin = value - sd, ymax = value + sd),
position = position_dodge(0.9), width = 0.2, alpha = 0.5, size = 1
) +
theme_bw()
}
p1 <- subset(data, !facet) |> plot_fun() + scale_y_continuous(limits = c(0, 1))
p2 <- subset(data, facet) |> plot_fun() + scale_y_continuous(limits = c(-15, 15), position = "right")
p1 + p2 +
plot_layout(guides = "collect", width = c(3, 1))
A second but similar option would be to use ggh4x which via ggh4x::facetted_pos_scales allows to set the limits for facet panels individually. One drawback, the panels have the same width. (I failed in making this approach work with facet_grid and space="free")
library(ggplot2)
library(ggh4x)
data$facet <- data$index %in% "PBIAS"
ggplot(data, aes(x = index, y = value, fill = period)) +
geom_bar(position = "dodge", stat = "identity") +
geom_errorbar(aes(ymin = value - sd, ymax = value + sd),
position = position_dodge(0.9), width = 0.2, alpha = 0.5, size = 1
) +
facet_wrap(~facet, scales = "free") +
facetted_pos_scales(
y = list(
facet ~ scale_y_continuous(limits = c(-15, 15), position = "right"),
!facet ~ scale_y_continuous(limits = c(0, 1), position = "left")
)
) +
theme_bw() +
theme(strip.text.x = element_blank())
I want my bars to correspond to a legend and to have them colored with my own color, not the default one.
# library
library(ggplot2)
# create a dataset
specie <- c(rep("IFNg_WNH", 2) ,
rep("IFNg_AA", 2),
rep("IL1b_WNH", 2),
rep("IL1b_AA", 2),
rep("IL6_WNH", 2),
rep("IL6_AA", 2)
)
condition <- rep(c("down", "up"), 6)
value <- c(452,216,
348,327,
207,61,
75,53,
177,191,
379,318)
data <- data.frame(specie,condition,value)
data
# Grouped
p <- ggplot(data, aes(fill=condition, y=value, x=specie)) +
geom_bar(position="dodge", stat="identity")
z <- p+labs(y = "Number of genes", x = "Cytokines")+
theme_classic()+
theme(plot.title = element_text(hjust = 0.5))+
theme(axis.line=element_line(size=1))+
scale_y_continuous(expand = c(0, 0), limits = c(0, NA))+
scale_fill_discrete(labels=c('up', 'down'))
z
Once I add
z + scale_fill_manual(values=c('#eb4034','#0a0a0a'))
The color is changing but the legend reverts to the wrong one. What is happening?
Adding type to scale_fill_discrete
type: One of the following:
• A character vector of color codes.
• A list of character vectors of color codes.
• A function that returns a discrete colour/fill scale
ggplot(data, aes(fill=condition, y=value, x=specie)) +
geom_bar(position="dodge", stat="identity") +
labs(y = "Number of genes", x = "Cytokines") +
theme_classic() +
theme(plot.title = element_text(hjust = 0.5)) +
scale_y_continuous(expand = c(0, 0), limits = c(0, NA)) +
scale_fill_discrete(labels=c('up', 'down'), type=c('#eb4034','#0a0a0a'))
There are a couple issues here. First, scale_fill_manual() essentially “overwrites” scale_fill_discrete(). Instead, use just one scale_fill_*() call including all relevant arguments:
library(ggplot2)
p +
labs(y = "Number of genes", x = "Cytokines")+
theme_classic()+
theme(plot.title = element_text(hjust = 0.5))+
theme(axis.line=element_line(size=1))+
scale_y_continuous(expand = c(0, 0), limits = c(0, NA))+
scale_fill_manual(labels = c('up', 'down'), values=c('#eb4034', '#0a0a0a'))
However - currently, the labels argument is in effect recoding your data, so that "up" values are labeled "down" and vice versa, which I assume isn’t what you want. My best guess is you’re actually trying to change the order the labels appear in the legend. If so, you can change the factor levels of condition to the order you want:
data$condition <- factor(data$condition, c("up", "down"))
p <- ggplot(data, aes(fill=condition, y=value, x=specie)) +
geom_bar(position="dodge", stat="identity")
p +
labs(y = "Number of genes", x = "Cytokines")+
theme_classic()+
theme(plot.title = element_text(hjust = 0.5))+
theme(axis.line=element_line(size=1))+
scale_y_continuous(expand = c(0, 0), limits = c(0, NA))+
scale_fill_manual(values=c('#eb4034', '#0a0a0a'))
I am making a set of scorecards where I am generating a set of graphs that show the distribution of responses from a survey and also where the response for a specific company falls. I need to modify the formatting of a graph, a stacked barchart, and add a few features I’ve outlined below. I’ve already spent a few hours getting my chart to where it is now and would appreciate your help with the features I outline below.
Data is
Data<-data.frame(Reviewed = c("Annually", "Annually", "Hourly", "Monthly", "Weekly","Monthly","Weekly","Other","Other","Monthly","Weekly"),Company=c("a","b","c","d","e","f","g","h","i","j","k"),Question="Q1")
So far I’ve developed this
ggplot(Data, aes(x="Question", fill=Reviewed)) + geom_bar(position='fill' ) +
coord_flip()
I would like to do the following:
Order the variables so they are arranged on plot as follows: Annually,Monthly,Weekly,Hourly,Other
Express the y axis in terms of percent. I.e. 0.25 turns into 25%
Move y-axis directly underneath the bar.
Remove the legend but move the terms underneath the respective part of the graph on a diagonal slant.
Add a black line that cuts down the 50% mark
Add a dot in at the midpoint of the stack for the value of company “e”.
Remove gray background
This is what I'm hoping the finished graph will look like.
There's a lot to unpack here, so I'll break it down bit by bit:
Order the variables so they are arranged on plot as follows: Annually,Monthly,Weekly,Hourly,Other
Assign "Reviewed" as an ordered factor. I'm reversing the order here since it wants to plot the "lowest" factor first (to the left).
Data$Reviewed <- factor(Data$Reviewed,
levels = rev(c('Annually', 'Monthly', 'Weekly', 'Hourly', 'Other')),
ordered = T)
ggplot(Data, aes(x="Question", fill=Reviewed)) + geom_bar(position='fill' ) +
coord_flip()
Express the y axis in terms of percent. I.e. 0.25 turns into 25%
Use scale_y_continuous(labels = scales::percent) to adjust the labels. I believe that the scales was pulled in when you installed ggplot2.
ggplot(Data, aes(x="Question", fill=Reviewed)) +
geom_bar(position = 'fill') +
scale_y_continuous(labels = scales::percent) +
coord_flip()
Move y-axis directly underneath the bar.
Remove gray background
These are done all at once by adding expand = F to coord_flip.
ggplot(Data, aes(x="Question", fill=Reviewed)) +
geom_bar(position = 'fill') +
scale_y_continuous(labels = scales::percent) +
coord_flip(expand = F)
Remove the legend...
Add theme(legend.position = 'none').
ggplot(Data, aes(x="Question", fill=Reviewed)) +
geom_bar(position = 'fill') +
scale_y_continuous(labels = scales::percent) +
coord_flip(expand = F) +
theme(legend.position = 'none')
but move the terms underneath the respective part of the graph on a diagonal slant.
This is tougher and takes a good amount of fiddling.
Use geom_text to make the labels
Calculate the position along the bar using the 'count' stat
Move the labels to the bottom of the plot by providing a fake x coordinate
Align the labels in the center of the bars using position_stack, and make them abut the x axis using hjust.
Add angle.
Use clip = 'off' in coord_flip to make sure that these values are not cut out since they're outside the plotting area.
Fiddle with the x limits to crop out empty plotting area.
Adjust the plot margin in theme to make sure everything can be seen.
ggplot(Data, aes(x="Question", fill=Reviewed)) +
geom_bar(position = 'fill') +
geom_text(aes(label = Reviewed, x = 0.45,
y = stat(..count../sum(..count..))), stat = 'count',
position = position_stack(0.5),
hjust = 0,
angle = 45) +
scale_y_continuous(labels = scales::percent) +
coord_flip(xlim = c(0.555, 1.4), clip = 'off',expand = F) +
theme(plot.margin = margin(0, 0, 35, 10),
legend.position = 'none')
Add a black line that cuts down the 50% mark
Use geom_hline(yintercept = 0.5); remember that it's a "horizontal" line since the coordinates are flipped.
ggplot(Data, aes(x="Question", fill=Reviewed)) +
geom_bar(position = 'fill') +
geom_text(aes(label = Reviewed, x = 0.45,
y = stat(..count../sum(..count..))), stat = 'count',
position = position_stack(0.5),
hjust = 0,
angle = 45) +
geom_hline(yintercept = 0.5) +
scale_y_continuous(labels = scales::percent) +
coord_flip(xlim = c(0.555, 1.4), clip = 'off',expand = F) +
theme(plot.margin = margin(0, 0, 20, 10),
legend.position = 'none')
Add a dot in at the midpoint of the stack for the value of company “e”.
This is pretty hack-y. Using the same y values as in geom_text, use geom_point to plot a point for every value of Reviewed, then use position_stack(0.5) to nudge them to the center of the bar. Then use scale_color_manual to only color "Weekly" values (which is the corresponding value of Reviewed for Company "e"). I'm sure there's a way to do this more programmatically.
ggplot(Data, aes(x="Question", fill=Reviewed)) +
geom_bar(position = 'fill') +
geom_text(aes(label = Reviewed, x = 0.45,
y = stat(..count../sum(..count..))), stat = 'count',
position = position_stack(0.5),
hjust = 0,
angle = 45) +
geom_hline(yintercept = 0.5) +
geom_point(aes(y = stat(..count../sum(..count..)),
color = Reviewed), stat = 'count',
position = position_stack(0.5), size = 5) +
scale_color_manual(values = 'black', limits = 'Weekly') +
scale_y_continuous(labels = scales::percent) +
coord_flip(xlim = c(0.555, 1.4), clip = 'off',expand = F) +
theme(plot.margin = margin(0, 0, 20, 10),
legend.position = 'none')
This is what I'm hoping the finished graph will look like.
Prettying things up:
ggplot(Data, aes(x="Question", fill = Reviewed)) +
geom_bar(position = 'fill') +
geom_text(aes(label = Reviewed, x = 0.45,
y = stat(..count../sum(..count..))), stat = 'count',
position = position_stack(0.5),
hjust = 0,
angle = 45) +
geom_hline(yintercept = 0.5) +
geom_point(aes(y = stat(..count../sum(..count..)),
color = Reviewed), stat = 'count',
position = position_stack(0.5), size = 5) +
scale_color_manual(values = 'black', limits = 'Weekly') +
scale_y_continuous(labels = scales::percent) +
coord_flip(xlim = c(0.555, 1.4), clip = 'off', expand = F) +
labs(x = NULL, y = NULL) +
theme_minimal() +
theme(plot.margin = margin(0, 0, 35, 10),
legend.position = 'none')
I have a bar plot that has 12 x values. Using this code I get the plot I want except for the x-axis labels.
p <- ggplot(data=df.mean, aes(x=stock_name, y=invest_amnt, fill=trend_id)) +
geom_bar(stat="identity", position=position_dodge()) +
geom_errorbar(aes(ymin=invest_amnt-ic, ymax=invest_amnt+ic), width=.2,
position=position_dodge(.9))
p + scale_fill_brewer(palette="Paired") + theme_minimal() +
theme(text = element_text(size=12, hjust = 0.5, family="Times")) +
theme_stata() + scale_color_stata()
Instead of displaying all 12 values on the x-axis I want to determine the labels by myself and only display 4.
I adjusted the code like this
p <- ggplot(data=df.mean, aes(x=stock_name, y=invest_amnt, fill=trend_id)) +
geom_bar(stat="identity", position=position_dodge()) +
geom_errorbar(aes(ymin=invest_amnt-ic, ymax=invest_amnt+ic), width=.2,
position=position_dodge(.9)) +
scale_x_discrete( labels=c("UP\nDOWN", "DOWN\nUP", "STRAIGHT\nGAIN", "STRAIGHT\nLOSS")) +
scale_fill_discrete(name = "Trend", labels = c("negative", "flat", "positive"))
p + scale_fill_brewer(palette="Paired") + theme_minimal() +
theme(text = element_text(size=12, hjust = 0.5, family="Times")) +
theme_stata() + scale_color_stata()
Unfortunately, I get my 4 labels but also 8 NAs. I would like my 4 labels to be evenly spread on my x-axis. Since my labels are factors I do not know how to apply break here.
I've generated some sample data...hope I've understood the situation correctly.
This seems to work, i.e. inserts breaks at the specified locations on a barplot, using the specified labels.
library(tidyverse)
df <- tribble(~x, ~y,
'cat',10,
'dog', 20,
'rabbit', 30,
'fox', 30)
df <- df %>%
mutate(x = factor(x))
df %>% ggplot(aes(x,y))+
geom_bar(stat = 'identity') +
scale_x_discrete(breaks = c('cat','fox'), labels = c('pig', 'hen'))
I need to combine the boxplot with the histogram using ggplot2. So far I have this code.
library(dplyr)
library(ggplot2)
data(mtcars)
dat <- mtcars %>% dplyr::select(carb, wt) %>%
dplyr::group_by(carb) %>% dplyr::mutate(mean_wt = mean(wt), carb_count = n())
plot<-ggplot(data=mtcars, aes(x=carb, y=..count..)) +
geom_histogram(alpha=0.3, position="identity", lwd=0.2,binwidth=1)+
theme_bw()+
theme(panel.border = element_rect(colour = "black", fill=NA, size=0.7))+
geom_text(data=aggregate(mean_wt~carb+carb_count,dat,mean), aes(carb, carb_count+0.5, label=round(mean_wt,1)), color="black")
plot + geom_boxplot(data = mtcars,mapping = aes(x = carb, y = 6*wt,group=carb),
color="black", fill="red", alpha=0.2,width=0.1,outlier.shape = NA)+
scale_y_continuous(name = "Count",
sec.axis = sec_axis(~./6, name = "Weight"))
This results in
However, I dont want the secondary y axis to be the same length of primary y axis. I want the secondary y axis to be smaller and on the top right corner only. Lets say secondary y axis should scale between 20-30 of primary y axis and the box plot should also scale with the axis.
Can anyone help me with this?
Here's one approach, where I adjusted the secondary axis formula and tweaked the way it's labeled. (EDIT: adjusted to make boxplots bigger, per OP comment.)
plot + geom_boxplot(data = mtcars,
# Adj'd scaling so each 1 wt = 2.5 count
aes(x = carb, y = (wt*2.5)+10,group=carb),
color="black", fill="red", alpha=0.2,
width=0.5, outlier.shape = NA)+ # Wider width
scale_y_continuous(name = "Count", # Adj'd labels to limit left to 0, 5, 10
breaks = 5*0:5, labels = c(5*0:2, rep("", 3)),
# Adj'd scaling to match the wt scaling
sec.axis = sec_axis(~(.-10)/2.5, name = "Weight",
breaks = c(0:5))) +
theme(axis.title.y.left = element_text(hjust = 0.15, vjust = 1),
axis.title.y.right = element_text(hjust = 0.15, vjust = 1))
You might also consider an alternative using the patchwork package, coincidentally written by the same developer who implemented secondary scales in ggplot2...
# Alternative solution using patchwork
library(patchwork)
plot2 <- ggplot(data=mtcars, aes(x=carb, y=..count..)) +
theme_bw()+
theme(panel.border = element_rect(colour = "black", fill=NA, size=0.7))+
geom_boxplot(data = mtcars,
aes(x = carb, y = wt, group=carb),
color="black", fill="red", alpha=0.2,width=0.1,outlier.shape = NA) +
scale_y_continuous(name = "Weight") +
scale_x_continuous(labels = NULL, name = NULL,
expand = c(0, 0.85), breaks = c(2,4,6,8))
plot2 + plot + plot_layout(nrow = 2, heights = c(1,3)) +
labs(x=NULL)