for loop to create multiple ggplots from single data frame

Question

I've been trying to create a for loop that will go through my data frame and make the same plot for each of the individual IDs I have. Here's some sample data:

TJID1 <- c("TJ22", "TJ22", "TJ23", "TJ23", "TJ23", "TJ24", "TJ24")
Day <- c("2005-11-22", "2005-11-23", "2006-12-01", "2006-12-02", "2006-12-03","2005-07-08", "2005-07-08")
Mean.Depth <- c (2, 2, 3, 4, 5, 6, 6)
SE.Depth <- c(1, 1, 2, 2, 1, 2, 2)
sample  <- cbind(TJID1, Day, Mean.Depth, SE.Depth)
sample <- as.data.frame(sample)

I have each individual as a different TJ number, and then for each TJ they have different daily depth means. The days change for each individual. I have been able to subset the main data frame by each individual TJ TJ22 <- sample [sample$TJID1 == "TJ22", ]. My code from the plot is then (using the subsetted data frame):

DailyMeans_TJ22 <- ggplot(TJ22, aes(x=Day, y=Mean.Depth))+
  geom_point()+
  geom_line()+
  geom_errorbar(aes(ymin=Mean.Depth-1.96*SE.Depth, ymax=Mean.Depth+1.96*SE.Depth), width = 0.5, col="red")

I want to create a for loop that just cycles through each individual and makes the same plot. This is what I have so far:

var_list = combn(names(sample) [3:4], 3, simplify=FALSE)
plot_list = list()
for (i in unique (sample$TJID1)){
     TJ <- sample[sample$TJID1== i,]

     p = ggplot(TJ, aes_string(x=var_list[[i]][1], y=var_list[[i]][2])) +
    geom_point()+
    geom_line()

 plot_list[[i]] = p
 }

But that literally just gives me nothing. Any help is appreciated!

you have to print(p). Why not just use facet_wrap(~TJID1) instead of the loop? — Neal Fultz
– Neal Fultz, Commented Aug 1, 2017 at 4:05

www · Accepted Answer · 2017-08-01 05:00:50Z

1

Here is an idea. We can design a function to subset sample and then create and return a plot. After that, we used lapply to loop through the unique values in TJID1.

Please notice that in your original sample data frame, those numeric columns are expressed as factor. I changed the way to create the sample data frame to fix that. One final note. sample is a bad name because there is a function in R called sample, which causes confusion. Please name your data frame using other names that do not match other function names in the future.

# Load package
library(ggplot2)

# Create example data frame
sample <- data.frame(TJID1 = c("TJ22", "TJ22", "TJ23", "TJ23", "TJ23", "TJ24", "TJ24"),
                     Day = c("2005-11-22", "2005-11-23", "2006-12-01", "2006-12-02", "2006-12-03","2005-07-08", "2005-07-08"),
                     Mean.Depth = c (2, 2, 3, 4, 5, 6, 6),
                     SE.Depth = c(1, 1, 2, 2, 1, 2, 2),
                     stringsAsFactors = FALSE)

# Design a function
gg_fun <- function(parameter, dt){

  p <- ggplot(dt[dt$TJID1 == parameter, ], aes(x=Day, y=Mean.Depth))+
    geom_point()+
    geom_line()+
    geom_errorbar(aes(ymin=Mean.Depth-1.96*SE.Depth, ymax=Mean.Depth+1.96*SE.Depth), width = 0.5, col="red") +
    ggtitle(parameter)

  return(p)
}

# Apply the function
plot_list <- lapply(unique(sample$TJID1), gg_fun, dt = sample)

edited Aug 1, 2017 at 5:00

answered Aug 1, 2017 at 4:27

www

39.3k12 gold badges52 silver badges93 bronze badges

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

2 Comments

Ally D Over a year ago

wow! thanks so much :) I was so focused on trying to use a loop that I didn't even think of a function. This worked great!

Ally D Over a year ago

Also, if I wanted to add to the function + ggtitle (), how could I get it to display the TJ number for each individual plot as the title?

Collectives™ on Stack Overflow

for loop to create multiple ggplots from single data frame

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related