Constructing a line graph using ggplot2

Question

I need to plot three lines (onto a single graph) each of which represents one lab team's data (two variables / team). Ideally, the graph should look aesthetically pleasing (hence the use of ggplot2!) yet similar in form to the line graphs shown below. I don't understand how to plot multiple lines onto a single graph using the gggplot2 library. My current knowledge of / skill with the ggplot2 library is low but I have included my fledgling efforts below.

http://www.harding.edu/fmccown/r/#linecharts

Edit: Each line is constructed from two vectors that look like this:

temp = c(4, 25, 50, 85, 100) 
enzyme_activity = c(0.543, 0.788, 0.990, 0.898, 0.882)

with the temp variable on the x-axis and each line a different color so they can be differentiated.

Edit2:

amyA = c(0.091, 0.147, 0.202, 0.236, 0.074)
temp = c(4, 23, 37, 65, 100)
df = data.frame(temp, amyA)

ggplot(df, aes(x = temp, y = amyA, col = 'blue')) + geom_line()

The code within the second edit does not generate a blue line and the legend is completely wrong. If I repeat the two ggplot calls with differing data, only one line is plotted.

helpful reference: cookbook-r.com/Graphs/Bar_and_line_graphs_(ggplot2) — Ricardo Saporta
– Ricardo Saporta, Commented Sep 25, 2013 at 18:05
What have you tried already, and what does not work, or what do you not understand? — Paul Hiemstra
– Paul Hiemstra, Commented Sep 25, 2013 at 18:06

dardisco · Accepted Answer · 2013-09-26 04:50:15Z

5

The key thing is to organize your data before plotting so as to have a factor column in the data frame indicating a separate line for each set of x, y values. For example:

set.seed(1)
df1 <- data.frame(e1 = sort(runif(5, 0.05, 0.25)),
                  e2 = sort(runif(5, 0.05, 0.25)),
                  e3 = sort(runif(5, 0.05, 0.25)),
                  t1 = sort(runif(5, 1, 100)),
                  t2 = sort(runif(5, 1, 100)),
                  t3 = sort(runif(5, 1, 100))
                  )
### reshape this to give a column indicating group
df2 <- with(df1,
        as.data.frame(cbind( c(t1, t2, t3),
                            c(e1, e2, e3),
                            rep(seq(3), each=5) )
                      ))
colnames(df2) <- c("temp","activity","team")
df2$team <- as.factor(df2$team)

Then

library(ggplot2)
ggplot(df2, aes(x=temp, y=activity, col=team)) + geom_line()

giving: enter image description here

edited Sep 26, 2013 at 4:50

answered Sep 25, 2013 at 22:52

dardisco

5,3042 gold badges42 silver badges54 bronze badges

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

2 Comments

CaitlinG Over a year ago

I attempted to use the code you posted and my graph looked nothing like your example. Did you define vectors that looked like the following? > e1 [1] 0.091 0.147 0.202 0.236 0.074 > e2 [1] 0.035 0.058 0.090 0.138 0.075 > e3 [1] 0.036 0.135 0.127 0.135 0.033 > > t1 [1] 4 23 37 65 100 > t2 [1] 4 23 37 65 100 > t3 [1] 4 23 37 65 100

dardisco Over a year ago

Sorry! I left out the as.factor line when I was pasting. It should work just as above now. If you use the same set.seed you should get the same vectors as above e.g. df1$e1 is 0.09033639 0.10310173 0.12442478 0.16457067 0.23164156

Paul Hiemstra · Accepted Answer · 2013-09-25 18:03:36Z

2

I'd do something like:

library(ggplot2)
ggplot(mtcars, aes(x = wt, y = mpg, color = as.factor(cyl))) + geom_line()

enter image description here

If you want more specific advice, I would suggest you expand your example, and include some example data, and provide more details regarding what the visualisation should tell.

answered Sep 25, 2013 at 18:03

Paul Hiemstra

61.2k12 gold badges146 silver badges151 bronze badges

1 Comment

CaitlinG Over a year ago

Hi Paul. Thanks. Please see my edited post for additional detail.

Collectives™ on Stack Overflow

Constructing a line graph using ggplot2

2 Answers 2

2 Comments

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related