I am trying to do an alluvial plot which compares ordination methods after clustering. The idea is to have 3 variables on the x axis, corresponding to PCA, PCoA and PLN Model.
I would like to "cut" each one of these into 4 parts (4 clusters given by my hierarchical clustering).
Here is my dataset :
data_alluvial <- read.table(text = "
Var1 Freq method
1 1 4 ACP
2 2 9 ACP
3 3 8 ACP
4 4 4 ACP
5 1 6 PLNPCA
6 2 7 PLNPCA
7 3 5 PLNPCA
8 4 7 PLNPCA
9 1 8 PCoA
10 2 4 PCoA
11 3 9 PCoA
12 4 4 PCoA", header = TRUE)
and here is what I tried :
ggplot(data = data_alluvial, aes(x = method, stratum = Var1, alluvium = Freq, label = Var1)) + geom_alluvium(aes(fill = Var1))
I get the message :
Data is not in a recognized alluvial form (see
help('alluvial-data')
for details).
I don't really get what is wrong with my dataset...can someone help me ?
Try this:
#convert to factor
data_alluvial$Var1 <- as.factor(data_alluvial$Var1)
ggplot(data = data_alluvial,
aes(x = method, y = Freq, alluvium = Var1,
fill = Var1, stratum = Var1)) +
geom_alluvium() +
geom_stratum()