Order the column of a plot based on the counts of level

Question

I have the following data frame:

ddat <- data.frame(Canonical_Hugo_Symbol=sample(letters[1:4], 20, replace = TRUE), ID.name=rep(c("0", "1"), each=10), allele.fraction =runif(20,min=0,max=1), Canonical_Variant_Classification =sample(letters[1:4], 20, replace = TRUE))

    Canonical_Hugo_Symbol ID.name allele.fraction Canonical_Variant_Classification
1                      b       0      0.47877015                                a
2                      a       0      0.98445203                                a
3                      d       0      0.91065285                                d
4                      b       0      0.93833143                                c
5                      d       0      0.53332525                                d
6                      a       0      0.14730869                                a
7                      b       0      0.71067695                                b
8                      a       0      0.46656093                                d
9                      d       0      0.64203393                                b
10                     a       0      0.48894393                                b 
11                     a       1      0.17165993                                a
12                     a       1      0.02641931                                c
13                     b       1      0.46169460                                b
14                     b       1      0.96254767                                c
15                     a       1      0.81565680                                d
16                     c       1      0.95940276                                d
17                     b       1      0.28574428                                c
18                     c       1      0.10000156                                c
19                     c       1      0.19250335                                c
20                     c       1      0.72370884                                b

I want to plot a bar plot ordered on the Canonical_Hugo_Symbol count.

I use the command :

ddat$count <- ave(as.numeric(ddat$Canonical_Hugo_Symbol), ddat$Canonical_Hugo_Symbol, FUN = length)
ddat <- ddat[order(ddat$count, decreasing = T),]
qplot(Canonical_Hugo_Symbol, data= ddat, fill= Canonical_Variant_Classification, geom="bar")

but the column in the plot are still not ordered.

How can I do that?

This is pretty much a duplicate question. See this [stackoverflow answer](http://stackoverflow.com/questions/3744178/ggplot2-sorting-a-plot) — BrodieG, Jan 13 '14 at 13:30

Sven Hohenstein · Answer 1 · 2014-01-13T13:51:36.517

1

You can use the function reorder:

qplot(reorder(Canonical_Hugo_Symbol, count), 
      data = ddat, fill = Canonical_Variant_Classification, geom = "bar")

enter image description here

If you want to sort the bars in decreasing order, you can use:

reorder(Canonical_Hugo_Symbol, -count).

edited Jan 13 '14 at 13:51

answered Jan 13 '14 at 13:42

Sven Hohenstein

80,497
17
145
168

In some way is not reordering the column in my case. I had tried before. Thanks! – user3186183 Jan 13 '14 at 14:14

score 0 · Accepted Answer · answered Jan 13 '14 at 13:46

Is this what you had in mind:

library(ggplot2)
set.seed(1)
ddat <- data.frame(Canonical_Hugo_Symbol=sample(letters[1:4], 20, replace = TRUE), ID.name=rep(c("0", "1"), each=10), allele.fraction =runif(20,min=0,max=1), Canonical_Variant_Classification =sample(letters[1:4], 20, replace = TRUE))
ddat$count <- ave(as.numeric(ddat$Canonical_Hugo_Symbol), 
                  ddat$Canonical_Hugo_Symbol, FUN = length)
ddat       <- ddat[order(ddat$count,ddat$Canonical_Hugo_Symbol, decreasing = T),]
ddat$Canonical_Hugo_Symbol <- factor(ddat$Canonical_Hugo_Symbol, 
                                     levels=unique(ddat$Canonical_Hugo_Symbol))
qplot(Canonical_Hugo_Symbol, data= ddat, 
      fill= Canonical_Variant_Classification, geom="bar")

Order the column of a plot based on the counts of level

2 Answers2