ggplot2 problem with Date format

Question

[luzglongoria]

Hi there, I'm trying to plot my data but I am not able to present dates as I want (by month and with the name of the month). First of all I show you part of my data:

Ind Date    Parasitemia       Genera              Age       Year
11  15/7/17 0              Haemoproteus          Adult      2017
26  23/8/16 0,000274329    Haemoproteus        Juveniles    2016
10  1/2/17  0,000395499    Plasmodium            Adult      2017
4   15/8/17 0              Plasmodium            Adult      2017
10  1/9/16  0,002482416    Plasmodium            Adult      2016

as you can see I have Individuals (the same individua) that have been analyzed over time (from February 2016 to May 2017). My goal is to plot a general graph like this one:

The code I have been using is this one:

ggplot(BigPlot, aes(x = Date, y = log(Parasitemia), col = Genera)) + geom_smooth() +
geom_point(aes(shape = Genera), size = 2) + ylab("Log10 Parasitemia") + xlab("Date") + 
facet_grid(Year~Age~Genera) +
scale_size("fixed_y") + 
theme(axis.text = element_text(size = 20), legend.text = element_text(size=20),
 axis.title.x = element_text(size = 20), axis.title.y = element_text(size = 20),
legend.title = element_text(size = 20), strip.text.x = element_text(size = 20))+
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_blank()) +
 theme(legend.position = "none")

BUT If I try to modify the dates in order to show every month (by using the package lubridate I got an eeror message: same code that before plus

 (scale_x_date(labels=date_format("%b %y")))

Then,

Error: Invalid input: date_trans works with objects of class Date only

I have also tried with

ggplot(BigPlot, aes(x = Date, y = log(Parasitemia), col = Genera)) + geom_smooth() +
geom_point(aes(shape = Genera), size = 2) + ylab("Log10 Parasitemia") + xlab("Date") +
facet_grid(Year~Age~Genera) + 
 scale_size("fixed_y") +
theme(axis.text = element_text(size = 20), legend.text = element_text(size=20), 
axis.title.x = element_text(size = 20), axis.title.y = element_text(size = 20), 
legend.title = element_text(size = 20), strip.text.x = element_text(size = 20))+
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_blank()) +
 (scale_x_date(breaks = date_breaks("1 month"),
labels = date_format("%m"))) +
geom_rect(data = BigPlot, 
xmin = as.Date("2016-02-01", "%Y-%m-%d"),
 xmax = as.Date("2017-09-15",  "%Y-%m-%d"),
ymin = -Inf,
ymax = Inf),
 fill = "gray",
 alpha = 0.5)

but I got the same error message:

Error: Invalid input: date_trans works with objects of class Date only

I guess I need to modify my original file somehow. I have tried to to this by :

newdata$event_Date <- as.Date(newdata$event_date)

but then,

Error in as.Date.default(newdata$event_date) : 
do not know how to convert 'newdata$event_date' to class “Date”

Any ideas?

[russhh]

Does BigPlot %>% dplyr::mutate(Date = lubridate::dmy(Date)) %>% ggplot(..... help?

[luzglongoria]

No. It gives me this error:

Error: Mapping should be created with aes() oraes_()`. In addition: Warning messages: 1: package ‘bindrcpp’ was built under R version 3.4.4 2: All formats failed to parse. No formats found.

[Russ]

In the code snippet using as.Date you refer to the event_date column - how are the dates formatted in this column? is it the same as the "Date" column in the original data you showed us? If so, have a look at the as.Date help file:

"A character string. If not specified, it will try "%Y-%m-%d" then "%Y/%m/%d" on the first non-NA element, and give an error if neither works. Otherwise, the processing is via strptime"

Your date appears to be formatted as "%d/%m/%Y" - try specifying the format when running as.Date.

Answer 1

[cpad0112]

Data frame in OP is BigPlot for ggplot. However, as.Data is on some other data as I see in OP. Is this a typo? Code you are using, didn't throw any errors in producing the plot (except some minor errors): I didn't modify your code except for as.Date(Date) instead of Date.

library(dplyr)
test=read.csv("test.txt", sep = "\t", strip.white = T, stringsAsFactors = F) 
library(ggplot2)
ggplot(test, aes(x = as.Date(Date), y = log(Parasitemia), col = Genera)) + geom_smooth() +
    geom_point(aes(shape = Genera), size = 2) + ylab("Log10 Parasitemia") + xlab("Date") + 
    facet_grid(Year~Age~Genera) +
    scale_size("fixed_y") + 
    theme(axis.text = element_text(size = 20), legend.text = element_text(size=20),
          axis.title.x = element_text(size = 20), axis.title.y = element_text(size = 20),
          legend.title = element_text(size = 20), strip.text.x = element_text(size = 20))+
    theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_blank()) +
    theme(legend.position = "none")

To the values, i added some random numbers to get points.

> test
  Ind     Date Parasitemia       Genera       Age Year
1  11 15-07-17    20.00000 Haemoproteus     Adult 2017
2  26 23-08-16    10.00027 Haemoproteus Juveniles 2016
3  10  1-02-17    20.00040   Plasmodium     Adult 2017
4   4 15-08-17    30.00000   Plasmodium     Adult 2017
5  10  1-09-16    40.00248   Plasmodium     Adult 2016

[luzglongoria]

Yes, but my real data is bigger, I have more than 20 different dates and when I try to plot them I got only 3 or 4 different dates (and I want all the month to appear). that's why I am trying to include other commands from the package lubridate ("scale_x_date"). And when I include them I got the error message.