User input name to ggplot

Question

I am writing a function to plot heat map for users. In the following example, it plots the change of grade over time for different gender.

However, this is a special case. "Gender" may have other name like "Class". I will let user input their specific name and then make ggplot have the right label for each axis.

How do I modify my function "heatmap()" based on what I need?

sampledata <- matrix(c(1:60,1:60,rep(0:1,each=60),sample(1:3,120,replace = T)),ncol=3)
colnames(sampledata) <- c("Time","Gender","Grade")
sampledata <- data.frame(sampledata)
heatmap=function(sampledata,Gender)
{
sampledata$Time <- factor(sampledata$Time)
sampledata$Grade <- factor(sampledata$Grade)
sampledata$Gender <- factor(sampledata$Gender)
color_palette <- colorRampPalette(c("#31a354","#2c7fb8", "#fcbfb8","#f03b20"))(length((levels(factor(sampledata$Grade)))))
ggplot(data = sampledata) + geom_tile( aes(x = Time, y = Gender, fill = Grade))+scale_x_discrete(breaks = c("10","20","30","40","50"))+scale_fill_manual(values =color_palette,labels=c("0-1","1-2","2-3","3-4","4-5","5-6",">6"))+  theme_bw()+scale_y_discrete(labels=c("Female","Male"))
}

I think you need aes_string. Take a look at this and this

Tung
– Tung

2016-06-21 18:24:38 +00:00
Commented Jun 21, 2016 at 18:24 — Tung
– Tung, Commented Jun 21, 2016 at 18:24

thepule · Accepted Answer · 2016-06-22 21:46:27Z

3

The easiest solution is redefining the function using aes_string. When the function is called, you need to pass it the name of the column you want to use as a string.

heatmap=function(sampledata,y)
{
        sampledata$Time <- factor(sampledata$Time)
        sampledata$Grade <- factor(sampledata$Grade)
        sampledata$new_var <- factor(sampledata[,y])
        color_palette <- colorRampPalette(c("#31a354","#2c7fb8", "#fcbfb8","#f03b20"))(length((levels(factor(sampledata$Grade)))))
        ggplot(data = sampledata) + geom_tile( aes_string(x = "Time", y = "new_var", fill = "Grade"))+scale_x_discrete(breaks = c("10","20","30","40","50"))+scale_fill_manual(values =color_palette,labels=c("0-1","1-2","2-3","3-4","4-5","5-6",">6"))+  theme_bw()+scale_y_discrete(labels=c("Female","Male")) + ylab(y)
}


# Below an example of how you call the newly defined function
heatmap(sampledata, "Gender")

Alternatively if you want to retain the quote free syntax, there is a slightly more complex solution:

heatmap=function(sampledata,y)
{
        arguments <- as.list(match.call())
        axis_label <- deparse(substitute(y))
        y = eval(arguments$y, sampledata)

        sampledata$Time <- factor(sampledata$Time)
        sampledata$Grade <- factor(sampledata$Grade)
        sampledata$y <- factor(y)
        color_palette <- colorRampPalette(c("#31a354","#2c7fb8", "#fcbfb8","#f03b20"))(length((levels(factor(sampledata$Grade)))))
        ggplot(data = sampledata) + geom_tile( aes(x = Time, y = y, fill = Grade))+scale_x_discrete(breaks = c("10","20","30","40","50"))+scale_fill_manual(values =color_palette,labels=c("0-1","1-2","2-3","3-4","4-5","5-6",">6"))+  theme_bw()+scale_y_discrete(labels=c("Female","Male")) + ylab(axis_label)
}

# Below an example of how you call the newly defined function
heatmap(sampledata, Gender)

edited Jun 22, 2016 at 21:46

answered Jun 21, 2016 at 18:38

thepule

1,7611 gold badge12 silver badges22 bronze badges

Sign up to request clarification or add additional context in comments.

5 Comments

Yukun Over a year ago

Thanks a lot! Just one more question: in the second function, we are still using 'sampledata$Gender <- factor(sampledata$Gender)' which includes Gender. It does't work if I put 'sampledata$y <- factor(sampledata$y)'. How to solve this part?

thepule Over a year ago

Yeah I forgot to change that part. I have edited both codes to make them more general. Now everything should work.

Yukun Over a year ago

What if I want the label of y axis to be "Gender" instead of "new_var" and "y". (Gender is consistent with the user input, it can also be other names)

thepule Over a year ago

I have edited the code. Now the y label will have the same name as the variable you are passing to the function. In the first function you simply pass the variable name to ylab, since it is already a string. In the second there is a bit of extra code at the beginning to extract the name. I want more brownie points for this :P

Yukun Over a year ago

Thanks a lot! That's really helpful. I have been searching around but couldn't figure it out. Now it works perfectly!

Collectives™ on Stack Overflow

User input name to ggplot

1 Answer 1

5 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

5 Comments

Your Answer

Sign up or log in

Post as a guest

Related