CODEWITHSUNDEEP
rggplot2heatmapgeom-tile
Zeki Akyol
Zeki Akyol

Reputation: 330

Reorder rows according to two column values in geom_tile() heatmap

library(tidyverse)

df <- data.frame(country = c("USA", "USA", "USA", 
                             "UK", "UK", "UK", 
                             "Japan", "Japan", "Japan", 
                             "Germany", "Germany", "Germany",
                             "Sweden","Sweden","Sweden",
                             "Norway","Norway","Norway"),
                 year = c(2000, 2001, 2002, 
                          2000, 2001, 2002, 
                          2000, 2001, 2002, 
                          2000, 2001, 2002, 
                          2000, 2001, 2002, 
                          2000, 2001, 2002),
                 value = c(0,1,1,
                           1,1,1,
                           0,0,0,
                           0,0,1,
                           0,1,1,
                           0,1,0))

ggplot(df, aes(year, country, fill = factor(value)))+
  geom_tile(color = "white", size = 1)+
  theme_minimal()

Raw Plot

This is the data, code, and raw plot that I get. Is there a way that I can reorder country rows by their value values and year at the same time?

My goal: country rows will be ordered by their first value=1 in years.

Problem: because I also want legend to be factor, not continuous. This creates problem in ordering.

Ideal order:

Upvotes: 0

Views: 764

Answers (1)

MarBlo
MarBlo

Reputation: 4524

You can use fct_relevel from the package forcats, if this is what you are looking for.

library(tidyverse)

library(forcats)
ro <- c('Japan', 'Germany', 'Norway', 'USA','Sweden', 'UK')

df %>%
  mutate(country= fct_relevel(country, ro)) %>% 
ggplot(aes(year, country, fill = factor(value))) +
  geom_tile(color = "white", size = 1)+
  theme_minimal()

Upvotes: 1

Related Questions