How to loop through the columns and create different plots?

Question

I have a dataframe "A" with samples as rows and Genes as columns with gene expression values (RPKM).

        Gene1   Gene2           Gene3        Gene4        Gene5    Gene6
Sample1 0.02   0.038798682  0.1423662   2.778587067 0.471403939 18.93687655
Sample2 0      0.059227225  0.208765213 0.818810739 0.353671882 1.379027685
Sample3 0      0.052116384  0.230437735 2.535040249 0.504061015 9.773089223
Sample4 0.06   0.199264618  0.261100548 2.516963635 0.63659138  11.01441624
Sample5 0      0.123521916  0.273330986 2.751309388 0.623572499 34.0563519
Sample6 0      0.128767634  0.263491811 2.882878373 0.359322715 13.02402045
Sample7 0      0.080097356  0.234511372 3.568192768 0.386217698 9.068928569
Sample8 0      0.017421323  0.247775683 5.109428797 0.068760572 15.7490551
Sample9 0      2.10281137   0.401582013 8.202902242 0.140596724 60.25989178

To make a scatter plot showing correlation between Gene1 vs Gene5 and Gene6 I used following code:

library(tidyr)
library(ggplot2)
pdf("Gene1.pdf")
A %>% 
  gather(key = variable, value = values, Gene5:Gene6) %>% 
  ggplot(aes(Gene1, values)) + 
  geom_point() + 
  facet_grid(. ~ variable, scales = "free_x") + 
  geom_smooth(method = "lm", se = FALSE) + 
  scale_y_continuous(trans = "log2", labels = NULL, breaks = NULL) + 
  scale_x_continuous(trans = "log2", labels = NULL, breaks = NULL)
dev.off()

I also want to plot for

Gene2 vs Gene5 and Gene6
Gene3 vs Gene5 and Gene6
Gene4 vs Gene5 and Gene6

Ofcourse I can replace Gene1 in the code with other Gene to plot. But instead of manual replacement I would like to make a loop so that from Gene1 to Gene4 plots against Gene5 and Gene6 with different plots each saved in pdf with respective gene names.

Based on request this is an Update about dput(A):

structure(list(Gene1 = c(0.02, 0, 0, 0.06, 0, 0, 0, 0, 0), Gene2 = c(0.038798682, 
0.059227225, 0.052116384, 0.199264618, 0.123521916, 0.128767634, 
0.080097356, 0.017421323, 2.10281137), Gene3 = c(0.1423662, 0.208765213, 
0.230437735, 0.261100548, 0.273330986, 0.263491811, 0.234511372, 
0.247775683, 0.401582013), Gene4 = c(2.778587067, 0.818810739, 
2.535040249, 2.516963635, 2.751309388, 2.882878373, 3.568192768, 
5.109428797, 8.202902242), Gene5 = c(0.471403939, 0.353671882, 
0.504061015, 0.63659138, 0.623572499, 0.359322715, 0.386217698, 
0.068760572, 0.140596724), Gene6 = c(18.93687655, 1.379027685, 
9.773089223, 11.01441624, 34.0563519, 13.02402045, 9.068928569, 
15.7490551, 60.25989178)), class = "data.frame", row.names = c("Sample1", 
"Sample2", "Sample3", "Sample4", "Sample5", "Sample6", "Sample7", 
"Sample8", "Sample9"))

user2602640 · Accepted Answer · 2018-07-06 22:23:07Z

2

Would a loop work for you? I haven't tested (since you don't have the data as dput()), so may need some cleanup.

cols <- colnames(A)
cols <- cols[!cols %in% c("Gene5", "Gene6")]
for(i in cols){
    name <- paste(i, ".pdf", sep = "")
    id <- which(colnames(A) == i)
    # add a new column - this is the one accepting the "rotating" gene input
    A$Gene <- A[,id]

p <- A %>% 
        select(Gene, Gene5, Gene6) %>%
        gather(variable, values, Gene5:Gene6) %>% 
        ggplot(aes(Gene, values)) + 
        geom_point() + 
        facet_grid(. ~ variable, scales = "free_x") + 
        geom_smooth(method = "lm", se = FALSE) + 
        scale_y_continuous(trans = "log2", labels = NULL, breaks = NULL) + 
        scale_x_continuous(i, trans = "log2", labels = NULL, breaks = NULL)

    pdf(name)
    print(p)

     dev.off()
        }

edited Jul 6, 2018 at 22:23

answered Jul 6, 2018 at 21:46

user2602640

1,09710 silver badges28 bronze badges

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

9 Comments

beginner Over a year ago

nope this didn't work. And in for loop I don't want to mention names of Gene1 to Gene4.

user2602640 Over a year ago

Can you post the data as dput(A)? And what do you mean by not wanting to mention names? What would work better for you?

beginner Over a year ago

What I mean is in this data I need Gene1 to Gene4 against Gene5 and Gene6 as I mentioned in the question. But in my original data I have more than 50 genes which I need plot against Gene5 and Gene6

user2602640 Over a year ago

I made an edit to the answer - had a typo in the code. Give it another try? And if that doesn't work - please run dput(A) in R and post the output here, so that I can read it in easily and test the code :-)

user2602640 Over a year ago

Are all the 50 genes named as GeneX?

|

Onyambu · Accepted Answer · 2018-07-06 22:22:16Z

2

if you need an explicit for loop:

for (gene in paste0("Gene",1:4)){
 a= A %>% 
    gather(key = variable, value = values, Gene5:Gene6) %>% 
    ggplot(aes(get(gene), values)) + 
    geom_point() + 
    facet_grid(. ~ variable, scales = "free_x") + 
    geom_smooth(method = "lm", se = FALSE) + 
    scale_y_continuous(trans = "log2", labels = NULL, breaks = NULL) + 
    scale_x_continuous(trans = "log2", labels = NULL, breaks = NULL)

    print(a)

}

answered Jul 6, 2018 at 22:22

Onyambu

80.3k3 gold badges29 silver badges65 bronze badges

4 Comments

beginner Over a year ago

What if columns Gene1 to Gene4 are named with different names like TGF, DGTH, HJUY, DSER.

Onyambu Over a year ago

put all the names you want in a vector. eg s= c('TGF', 'DGTH', 'HJUY', 'DSER') and now do for(gene in s){….}

beginner Over a year ago

Thank you very much !! but u didn't show how to save the plots as pdfs

Onyambu Over a year ago

Everything else remains the same. I know you could do that. just include the pdf part in the loop ie pdf(gene) etc etc

Collectives™ on Stack Overflow

How to loop through the columns and create different plots?

2 Answers 2

9 Comments

4 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

9 Comments

4 Comments

Your Answer

Sign up or log in

Post as a guest

Related