since a long time I despair to straighten the label of the x-axis in my plot (ggplot2).
The challenge is that I have two geom_paths, each fetching the data from a different dataframe - I'm sure this will become a bit clearer in the code:
ggplot(data=dx, aes(x = year, y=en.x ))+
scale_y_continuous(breaks = scales::pretty_breaks(n = 2))+
geom_path(data=ps, aes(x, y, color = "Person 1", linetype="Person 1"), size=0.5)+
geom_path(data=pg, aes(x , y, color = "Person 2", linetype="Person 2"), size=0.5)+
scale_color_manual("",labels = c(Nutzer1, Nutzer2), values = c("Person 1" = Nutzer1Farbe, "Person 2" = Nutzer2Farbe)) +
scale_linetype_manual("",labels = c(Nutzer1, Nutzer2), values=c("Person 1"=Nutzer1Format, "Person 2"=Nutzer2Format))
The goal is, to Label the X-Axis with the years from the dataframe "dx", as shown in the aes-parameter. And it works! But only if you disable the geom_paths - shown below:
ggplot(data=dx, aes(x = year, y=en.x ))+
scale_y_continuous(breaks = scales::pretty_breaks(n = 2))+
#geom_path(data=ps, aes(x, y, color = "Person 1", linetype="Person 1"), size=0.5)+
#geom_path(data=pg, aes(x , y, color = "Person 2", linetype="Person 2"), size=0.5)+
scale_color_manual("",labels = c(Nutzer1, Nutzer2), values = c("Person 1" = Nutzer1Farbe, "Person 2" = Nutzer2Farbe)) +
scale_linetype_manual("",labels = c(Nutzer1, Nutzer2), values=c("Person 1"=Nutzer1Format, "Person 2"=Nutzer2Format))
I can't really understand why the paths destroy the labeling like this - it must be the aes parameters.
If someone has a solution for this, I would be extremely grateful!
This could be achieved like so:
Convert your original month variable to a date time before calling xspline. This way the interpolated date values could be easily converted back to datetime via e.g. lubridate::as_datetime.
besides that you could row bind your datasets which makes plotting a bit easier
library(ggplot2)
library(tidyr)
library(dplyr)
datengesamt <- datengesamt %>%
# Convert to datetime
mutate(month = as.POSIXct(month))
plot(1, 1)
ps <- xspline(datengesamt[,1], datengesamt[,2], 1, draw=FALSE)
pg <- xspline(datengesamt[,1], datengesamt[,3], 1, draw=FALSE)
pp <- list("Person 1" = data.frame(ps), "Person 2" = data.frame(pg)) %>%
bind_rows(.id = "id") %>%
mutate(x = lubridate::as_datetime(x))
ggplot(pp, aes(x, y, color = id, linetype = id)) +
scale_y_continuous(breaks = scales::pretty_breaks(n = 2)) +
geom_path(size=0.5) +
scale_x_datetime(date_labels = "%Y")
Related
I have to produce a scatter plot with many points.
I am already using the package "ggrepel" in order to avoid overlapping, but it sometimes still doesnt work. Is there a possibility to insert a linebreak into the labels (e.g. after a certain length)?
Thanks for help!
items <- c("A long description of the item",
"Another very long text descrbing the item",
"And finally another one ",
"This text exceeds the available space by far",
"Incredibly long text",
"Here we go with another one",
"A linebreak would help here",
"This has at least 20 characters")
items <- rep(items, 4)
df <- data.frame(
descs = items,
x = rnorm(n = length(items), mean = 2, sd = 2),
y = rnorm(n = length(items), mean = 2, sd = 2),
cat = as.factor(runif(length(items), min = 1, max = 6))
)
library(ggplot2)
library(tidyverse)
library(ggrepel)
df %>% ggplot(aes(x = x, y = y, color = cat)) + geom_point() +
#geom_text(aes(label = descs)) +
geom_text_repel(aes(label = descs)) +
theme_light() +
theme(legend.position="none")
rm(items)
rm(df)
You could use stringr::str_wrap to achieve line breaks at an appropriate point. For example, to limit lines to 20 characters, you can do:
df %>%
ggplot(aes(x = x, y = y, color = cat)) + geom_point() +
geom_text_repel(aes(label = stringr::str_wrap(descs, 20))) +
theme_light() +
theme(legend.position = "none")
I saw this interesting way of creating a publication timeline using geom_dumbell, so I created my own by first loading the libraries:
library(tidyverse)
library(ggalt)
library(ggrepel)
Entering in some data:
# create data frame
df <- data.frame(
paper = c("Paper 1", "Paper 1", "Paper 2", "Paper 2", "Paper 3", "Paper 3", "Paper 3", "Paper 3"),
round = c("first","revision","first","revision","first","first","first","first"),
submission_date = c("2019-05-23","2020-12-11", "2020-08-12","2020-10-28","2020-12-10","2020-12-11","2021-01-20","2021-01-22"),
journal_type = c("physics", "physics","physics","physics","chemistry","chemistry","chemistry","chemistry"),
journal = c("journal 1", "journal 1", "journal 2", "journal 2", "journal 3", "journal 4", "journal 5", "journal 6"),
status = c("Revise and Resubmit", "Waiting for Decision", "Revise and Resubmit", "Accepted", "Desk Reject","Desk Reject", "Desk Reject","Waiting for Decision"),
decision_date = c("2019-09-29", "2021-01-24", "2020-08-27", "2020-10-29", "2020-12-10","2021-01-05","2021-01-22","2021-01-24"),
step_complete = c("yes","no","yes","yes","yes","yes","yes", "no"),
duration_days = c(129,44,15,1,0,25,2,2))
# convert variables to dates
df$decision_date = as.Date(df$decision_date)
df$submission_date = as.Date(df$submission_date)
and, finally, creating my own basic timeline using this code:
ggplot(df, aes(x = submission_date, xend = decision_date,
y = paper, label = duration_days,
color = status)) +
geom_dumbbell(size = 1, size_x = 1) +
scale_color_manual(values=c("green", "red", "darkolivegreen4", "turquoise1")) +
labs(x=NULL, color = 'Status:',
y=NULL,
title="Timeline of Journal Submissions",
subtitle="Start date, decision date, and wait time (in days) for my papers.") +
#theme_ipsum_tw() +
ggrepel::geom_label_repel(nudge_y = -.25, show.legend = FALSE) +
theme(legend.position = 'top')
As you can see from the above image, I can't see the x-axis. Additionally, I'd like to put another aesthetic and legend on the right side for the journal, perhaps putting a different shape on each line. Any other bells and whistles using the above data would be fun, too. Thanks!
Ok, I finally found some time to figure this out with help from this terrific post. To start, let's load the revised list of packages:
library(tidyverse)
library(ggalt)
library(ggrepel)
library(gridExtra)
library(gtable)
library(grid)
For comprehensiveness, let's reload the data:
# create dataframe
df <- data.frame(
paper = c("Paper 1", "Paper 1", "Paper 2", "Paper 2", "Paper 3", "Paper 3", "Paper 3", "Paper 3"),
round = c("first","revision","first","revision","first","first","first","first"),
submission_date = c("2019-05-23","2020-12-11", "2020-08-12","2020-10-28","2020-12-10","2020-12-11","2021-01-20","2021-01-22"),
journal_type = c("physics", "physics","physics","physics","chemistry","chemistry","chemistry","chemistry"),
Journal = c("journal 1", "journal 1", "journal 2", "journal 2", "journal 3", "journal 4", "journal 5", "journal 6"),
status = c("Revise and Resubmit", "Waiting for Decision", "Revise and Resubmit", "Accepted", "Desk Reject","Desk Reject", "Desk Reject","Waiting for Decision"),
decision_date = c("2019-09-29", "2021-01-24", "2020-08-27", "2020-10-29", "2020-12-10","2021-01-05","2021-01-22","2021-01-24"),
step_complete = c("yes","no","yes","yes","yes","yes","yes", "no"),
duration_days = c(129,44,15,1,0,25,2,2)
)
# convert variables to dates
df$decision_date = as.Date(df$decision_date)
df$submission_date = as.Date(df$submission_date)
First, let's create the plot with the color legend and extract it. Because I want that legend to be on top, I make sure indicate that as my legend position. Note that I specify my preferred colors using the scale_color_manual argument:
# make plot with color legend
p1 <- ggplot(df, aes(x = submission_date, xend = decision_date,
y = paper, label = duration_days,
color = status)) +
geom_dumbbell(size = 1, size_x = 1) +
scale_color_manual(values=c("green", "red", "darkolivegreen4", "turquoise1")) +
labs(x=NULL, color = 'Status:',
y=NULL,
title="Timeline of Journal Submissions",
subtitle="Start date, decision date, and wait time (in days) for my papers.") +
ggrepel::geom_label_repel(nudge_y = -.25, show.legend = FALSE) +
theme(legend.position = 'top')
# Extract the color legend - leg1
leg1 <- gtable_filter(ggplot_gtable(ggplot_build(p1)), "guide-box")
Second, let's make the plot with the shape legend and extract it. Because I want this legend to be positioned on the right side, I don't need to even specify the legend position here. Note that I specify my preferred shapes using the scale_shape_manual argument:
# make plot with shape legend
p2 <- ggplot(df, aes(x = submission_date, xend = decision_date,
y = paper, label = duration_days,
shape = Journal)) +
geom_dumbbell(size = 1, size_x = 1) +
scale_shape_manual(values=c(15, 16, 17, 18, 19,25))+
labs(x=NULL, color = 'Status:',
y=NULL,
title="Timeline of Journal Submissions",
subtitle="Start date, decision date, and wait time (in days) for my papers.") +
ggrepel::geom_label_repel(nudge_y = -.25, show.legend = FALSE)
# Extract the shape legend - leg2
leg2 <- gtable_filter(ggplot_gtable(ggplot_build(p2)), "guide-box")
Third, let's make the full plot with no legend, specifying both the scale_color_manual and scale_shape_manual arguments as well as theme(legend.position = 'none'):
# make plot without legend
plot <- ggplot(df, aes(x = submission_date, xend = decision_date,
y = paper, label = duration_days,
color =status, shape = Journal)) +
geom_dumbbell(size = 1, size_x = 3) +
scale_color_manual(values=c("green", "red", "darkolivegreen4", "turquoise1")) +
scale_shape_manual(values=c(15, 16, 17, 18, 19,25))+
labs(x=NULL, color = 'Status:',
y=NULL,
title="Timeline of Journal Submissions",
subtitle="Start date, decision date, and wait time (in days) for my papers.") +
ggrepel::geom_label_repel(nudge_y = -.25, nudge_x = -5.25, show.legend = FALSE) +
theme(legend.position = 'none')
Fourth, let's arrange everything according to our liking:
# Arrange the three components (plot, leg1, leg2)
# The two legends are positioned outside the plot:
# one at the top and the other to the side.
plotNew <- arrangeGrob(leg1, plot,
heights = unit.c(leg1$height, unit(1, "npc") - leg1$height), ncol = 1)
plotNew <- arrangeGrob(plotNew, leg2,
widths = unit.c(unit(1, "npc") - leg2$width, leg2$width), nrow = 1)
Finally, plot and enjoy the final product:
grid.newpage()
grid.draw(plotNew)
As everyone will no doubt recognize, I relied very heavily on this post. However, I did change a few things, I tried be comprehensive with my explanation, and some others spent time trying to help, so I think it is still helpful to have this answer here.
Image of legend i would like to add to
I would like to know how can i add a simple observation number (n) in legend of this scatter plot in ggplot2
library(readr)
library(ggplot2)
library(dplyr)
All.mutations.no.inserts <- read_csv("All mutations no inserts.csv")
All.mutations.no.inserts$Fungicide <- factor(All.mutations.no.inserts$Fungicide, levels = c("SDHI 1",
"SDHI 2",
"SDHI 3",
"SDHI 4",
"SDHI 5",
"SDHI 6",
"SDHI 7",
"SDHI 8",
"SDHI 9",
"SDHI 10",
"SDHI 11",
"SDHI 12"))
All.mutations.no.inserts$SDH.mutation <- factor(All.mutations.no.inserts$`SDH.mutation`)
ggplot(All.mutations.no.inserts, aes(x = Fungicide, y = EC50, color = SDH.mutation)) +
geom_point(size = 4) +
scale_y_log10() +
theme_minimal() +
theme(axis.text.x=element_text(angle = -90, hjust = 0),
axis.title.x=element_blank())
How should i modify my code?
here's an example using dplyr. See the comments in the code.
library(dplyr)
library(ggplot2)
# sample data set
expand.grid(y = rnorm(20),
x = letters[1:5],
z = letters[6:10]) %>%
sample_frac(0.75) %>%
# add column n with counts for each group
add_count(z) %>%
# combine the group z and count n into one column
mutate(zn = paste0(z, ' (', n, ')')) %>%
# plot as you had
ggplot(aes(x, y, colour = zn)) +
geom_point() +
# rename the legend title
labs(colour = 'z (# obs)')
Created on 2019-02-06 by the reprex package (v0.2.1)
I am trying to highlight selected points based on their order statistics in a ggplot stat_qq output:
ydata <- data.frame(sample = c(rep("Sample 1", 100),
rep("Sample 2", 100),
rep("Sample 3", 100),
rep("Sample 4", 100)),
x=rnorm(400))
ydata <- ydata[order(ydata$sample, ydata$x),]
ydata$order <- 1:100
ggplot(ydata, aes(sample=x)) + stat_qq() + facet_wrap(~sample) + scale_x_continuous(breaks = -2:2, labels = function(x) paste0(x, " \n [",100 * signif(pnorm(-2:2, lower.tail=FALSE),2), "%]")) + theme_bw(base_size = 14, base_family = "sans") + labs(title = "Four Samples of 100 Observations From Normal Distribution",
caption = "4 Samples of n = 100 from Normal Distribution \nNumbers indicate order of value",
y = "Sample Value",
x = "Standard Deviation\n[%exceeding]") +
geom_text(data = ydata[ydata$order %in% c(2,16,50,84,98),], aes(x=qnorm(pnorm(x)), y=x, label = order), nudge_y = 1)
Which produced this:
Obviously my text notation is not highlighting the right points (the 2, 16, 50 84, 98th points). I wish I could also highlight the actual points in red. Would appreciate any suggestions.
You could calculate the qq values outside of ggplot and create a separate column to group the qq values into highlighted and not highlighted. Then you could plot them using geom_point with the grouping variable as a colour aesthetic. For example:
library(tidyverse)
# Generate data reproducibly
set.seed(2)
ydata <- data.frame(sample = c(rep("Sample 1", 100),
rep("Sample 2", 100),
rep("Sample 3", 100),
rep("Sample 4", 100)),
x=rnorm(400))
ydata <- ydata[order(ydata$sample, ydata$x),]
ydata$order <- 1:100
# Quantile indices to highlight
pts = c(2,16,50,84,98)
# Add qq values and grouping column to data frame and pipe into ggplot
# Use split and map to calculate the qq values separately for each Sample
split(ydata, ydata$sample) %>%
map_df(~ .x %>% mutate(xq = qqnorm(x, plot.it=FALSE)$x,
group = ifelse(order %in% pts, "A", "B"))) %>%
ggplot(aes(xq, x, colour=group)) +
geom_point(size=1) +
geom_text(aes(label=ifelse(group=="A", order, "")),
nudge_y=1, size=3) +
facet_wrap(~ sample) +
theme_bw(base_size = 14, base_family = "sans") +
scale_colour_manual(values=c("red", "black")) +
guides(colour=FALSE)
As an alternative, a quick hack would be to use ggplot_build to highlight specific points in your original plot (note though that something is not quite right with how you placed the labels relative to the highlighted points):
pts = rep(c(2,16,50,84,98), 4) + rep(seq(0,300,100), each=5)
# Assuming you've assigned your plot to the object p
pb = ggplot_build(p)
# Change point colors
pb$data[[1]][pts, "colour"] = "red"
# Change label colors
pb$data[[2]][["colour"]] = "red"
# Regenerate plot object
p = ggplot_gtable(pb)
plot(p)
You can apply stat="qq" to your geom_point and then use the colors assigned to new variable
ydata <- data.frame(sample = c(rep("Sample 1", 100),
rep("Sample 2", 100),
rep("Sample 3", 100),
rep("Sample 4", 100)),
x=rnorm(400))
ydata <- ydata[order(ydata$sample, ydata$x),]
ydata$order <- 1:100
ydata$highlight = ifelse(ydata$order %in% c(2,16,50,84,98), "#FF0000", "#000000")
ydata$order_txt = ifelse(ydata$order %in% c(2,16,50,84,98), ydata$order, "")
ggplot(ydata, aes(sample=x)) +
geom_point(color=ydata$highlight, stat="qq") +
geom_text(label=ydata$order_txt, stat="qq", nudge_y=1) +
facet_wrap(~sample) +
scale_x_continuous(breaks = -2:2, labels = function(x) paste0(x, " \n [",100 * signif(pnorm(-2:2, lower.tail=FALSE),2), "%]")) +
theme_bw(base_size = 14, base_family = "sans") +
labs(
title = "Four Samples of 100 Observations From Normal Distribution",
caption = "4 Samples of n = 100 from Normal Distribution \nNumbers indicate order of value",
y = "Sample Value",x = "Standard Deviation\n[%exceeding]")
I'm trying to create a series of bar charts (to be replicated for multiple sites) that highlight the difference between the main site and the satellite locations. I can come somewhat close using geom_point, but I'd like to have them represented as bar charts, where the bar starts at the lowest point, there are labels for the main site and satellite locations, as well as the difference between them. Here is some sample code and screenshots of what I have, and an idea of what I'd like it to look like.
library(ggplot2)
library(dplyr)
site <- c("Site A", "Main Site", "Site A", "Main Site", "Site A", "Main Site")
year <- c("2013", "2013", "2014", "2014","2015", "2015" )
value <- c(57, 74, 60, 50, 60, 68)
df <- data.frame (site, year, value)
df %>%
mutate (label = paste0(site, " (", value, ")")) %>%
ggplot (aes (x = year, y = value, group = site, colour = site)) +
geom_point (size = 0.5) +
scale_y_continuous(limits = c (0,100)) +
geom_text (aes(label = label))
Using the comment from #Gregor I managed to come up with something that will work. Probably isn't the most elegant solution but will work for now.
df %>%
spread(site, value) %>%
mutate (diff = SiteA - MainSite) %>%
mutate (AboveBelow = recode (diff," -100:-1 = 'Below';
0 = 'No Difference';
1:100 = 'Above'")) %>%
ggplot() +
scale_x_continuous(name = "Year", breaks = c (2013, 2014, 2015)) +
scale_y_continuous(name = "Percentage", limits = c(0,100)) +
geom_rect (aes (xmin = year - 0.33, xmax = year + 0.33, ymin = SiteA, ymax = MainSite, fill = AboveBelow)) +
geom_text (aes (x = year, y = ifelse (diff < 0, MainSite + 5, MainSite - 3), label = paste0("MainSite - ", MainSite))) +
geom_text (aes (x = year, y = ifelse (diff < 0, SiteA - 3, SiteA +5), label = paste0("SiteA - ", SiteA))) +
geom_text (aes (x = year, y = MainSite + (diff/2), label = diff)) +
scale_fill_manual(values = c("green", "red", "white" ))
Gives me this:
Following up on the comment from #gregor, you can try the below (note dcast is from reshape2 and the heavy use of dplyr
df %>%
dcast(year~site) %>%
mutate(midpt = (`Main Site` + `Site A`)/2
, dir = factor( (`Main Site` - `Site A`) > 0
, levels = c(FALSE,TRUE)
, labels = c("Negative", "Positive"))
, diff = abs(`Main Site` - `Site A`)) %>%
ggplot(aes(x = year
, y = midpt
, fill = dir
, height = diff)) +
geom_tile() +
scale_fill_manual(values = c("Positive" = "darkgreen"
, "Negative" = "red3"))
If you have more than 2 sites, you would likely want a more flexible solution, probably using dplyr directly.