CODEWITHSUNDEEP
rggplot2heatmap
cms72
cms72

Reputation: 177

Creating a second ggplot heatmap with the same row order as the first heatmap

I am trying to create a second heatmap in ggplot that would maintain the rows from the first heatmap so that I can compare the results.

I have a File_A that contains more than 1000 lines, similar to below

Animal Bloodtype Count
Horse  Opos      10
Horse  Apos      5
Horse  Bpos      4
Horse  ABpos     5
Horse  Oneg      6
Horse  Aneg      7
Horse  Bneg      9
Horse  ABneg     10
Horse  Unknown   10
Cat    Opos      12
Cat    Apos      15
Cat    Bpos      14
Cat    ABpos     15
Cat    Oneg      16
Cat    Aneg      17
Cat    Bneg      19
Cat    ABneg     14
Cat    Unknown   14
Dog    Opos      9
Dog    Apos      23
Dog    Bpos      12
Dog    ABpos     42
Dog    Oneg      45
Dog    Aneg      23
Dog    Bneg      45
Dog    ABneg     32
Dog    Unknown   32
Mouse  Opos      3
Mouse  Apos      4
Mouse  Bpos      5
Mouse  ABpos     3
Mouse  Oneg      6
Mouse  Aneg      8
Mouse  Bneg      8
Mouse  ABneg     20
Mouse  Unknown   20
Pig    Opos      19
Pig    Apos      13
Pig    Bpos      22
Pig    ABpos     32
Pig    Oneg      25
Pig    Aneg      13
Pig    Bneg      35
Pig    ABneg     22
Pig    Unknown   22

And a modified_File_A that also contains more than 1000 lines, similar to below:

Animal Bloodtype Count
Horse  O      310
Horse  A      35
Horse  B      234
Horse  AB     325
Horse  O      346
Horse  A      427
Horse  B      439
Horse  AB     410
Horse  Unknown   210
Cat    O      312
Cat    A      215
Cat    B      314
Cat    AB     415
Cat    O      316
Cat    A      617
Cat    B      419
Cat    AB     314
Cat    Unknown   214
Dog    O      239
Dog    A      223
Dog    B      312
Dog    AB     342
Dog    O      245
Dog    A      423
Dog    B      345
Dog    AB     532
Dog    Unknown   132
Mouse  O      13
Mouse  A      24
Mouse  B      65
Mouse  AB     33
Mouse  O      56
Mouse  A      48
Mouse  B      28
Mouse  AB     320
Mouse  Unknown   202
Pig    O      193
Pig    A      135
Pig    B      224
Pig    AB     321
Pig    O      252
Pig    A      323
Pig    B      352
Pig    AB     222
Pig    Unknown   222

This is the code I used to create the FIRST heatmap

csv_file<-read.csv("~/Documents/FileA.csv")

csv_file.s <- ddply(csv_file, .(Bloodtype), transform, rescale = scale(Count))

csv_file.s$Category <- csv_file.s$Bloodtype

levels(csv_file.s$Category) <- 
  list("Opos" = c("Opos"),
       "Apos" = c("Apos"),
       "Bpos" = c("Bpos"),
       "ABpos" = c("ABpos"),
       "Oneg" = c("Oneg"),
       "Aneg" = c("Aneg"),
       "Bneg" = c("Bneg"),
       "Oneg" = c("Oneg"),
       "Unknown" = c("Unknown"))

csv_file.s$rescaleoffset <- csv_file.s$rescale + 100*(as.numeric(as.factor(csv_file.s$Category))-1)
scalerange <- range(csv_file.s$rescale)
gradientends <- scalerange + rep(c(0,100,200), each=8)
colorends <- c("white", "Aquamarine4", "white", "yellow4", "white", "turquoise4","white","orange4", "white", "slategray4","white","seagreen4","white","purple4","white","red4","white","blue4")

ggplot(csv_file.s, aes(x = Bloodtype, y = Animal, fill = Bloodtype, alpha = Count)) +
  geom_tile(fill = "white", alpha = 1) +
  geom_tile() +
  scale_alpha_continuous(breaks = seq(0, max(csv_file.s$Count), length.out =
10),limits = c(0, NA))

Which arranges the y-axis as: Pig Mouse Horse Dog Cat

I am trying to use the same code to create a second heatmap but it doesnt maintain the y-axis order from the first heatmap..

csv_file<-read.csv("~/Documents/Modified_FileA.csv")

csv_file.s <- ddply(csv_file, .(Bloodtype), transform, rescale = scale(Count))

csv_file.s$Category <- csv_file.s$Bloodtype

levels(csv_file.s$Category) <- 
list("O" = c("O"),
     "A" = c("A"),
     "B" = c("B"),
     "AB" = c("AB"),
     "Unknown" = c("Unknown"))

csv_file.s$rescaleoffset <- csv_file.s$rescale + 100*(as.numeric(as.factor(csv_file.s$Category))-1)
scalerange <- range(csv_file.s$rescale)
gradientends <- scalerange + rep(c(0,100,200), each=4)
colorends <- c("white", "Aquamarine4", "white", "yellow4", "white", "turquoise4","white","orange4","white","blue4")

ggplot(csv_file.s, aes(x = Bloodtype, y = Animal, fill = Bloodtype, alpha = Count)) +
  geom_tile(fill = "white", alpha = 1) +
  geom_tile() +
  scale_alpha_continuous(breaks = seq(0, max(csv_file.s$Count), length.out =
10),limits = c(0, NA))

I'm not sure if there is a way I can assign the y-axis order from first heatmap to a variable so I can introduce it to second heatmap as a level?

Upvotes: 0

Views: 197

Answers (1)

bischrob
bischrob

Reputation: 584

I think you might just have some typos in your code. The ggplot is using df as a variable, not the csv_file.s you have in your code. I tried your code and the y-axis was the same for both datasets you provided once I updated the variables. It's a good idea to restart R when testing reproducible code and make sure to include everything, including your library calls. Good luck

Upvotes: 1

Related Questions