How to change color of stacked bars after certain value in Swimmer ggplot in R-CodePudding

I'm wondering how can I change the color of the stacked bar after 2 days (geometric vertical line)?

As a result, I want to get the figure marked with yellow color.

Here is an example from my data frame with the indicator (marked in yellow) if the subject continues after two days.


 ggplot(aes(subjectID, endTime))   
   scale_fill_manual(values = c("red", "blue"))  
      geom_bar(stat="identity", aes(fill=factor(stage)))   
   geom_point(data=df, aes(subjectID, endTime, colour=responseType, shape=responseType),  size=6)  
   coord_flip()  
   scale_colour_manual(values=c('#20A387FF','#2D708EFF'))   
   scale_y_continuous(limits=c(-0.2, 20), breaks=0:20)   
      labs(fill="Treatment status",  # Add labels
        x="Subject ID ", y="Days since hospital admission",
        title="Patient trajectories",
        caption="Hypothetical example")  
   geom_hline(yintercept=2, linetype="dashed",  size=1)  # , col=c('DC0000B2')
   theme(plot.title = element_text(hjust = 0.5), # Put title in the middle of plot
         plot.caption = element_text(size=7, hjust=0))  # Make caption size smaller
   theme_minimal()

CodePudding user response：

I think I would do this with geom_rect

ggplot(df, aes(subjectID, endTime))   
  scale_fill_manual(values = c("red", "blue"))   
  geom_col(aes(fill = factor(stage)))   
  geom_rect(aes(ymin = 2, ymax = endTime, xmin = as.numeric(subjectID) - 0.45,
                xmax = as.numeric(subjectID)   0.45), fill = '#fff200',
            data = df[df$isContinued,])  
  geom_point(aes(colour = responseType, shape = responseType),  size = 6)  
  coord_flip()  
  scale_colour_manual(values = c('#20A387FF','#2D708EFF'))   
  scale_y_continuous(limits = c(-0.2, 20), breaks = 0:20)   
  labs(fill    = "Treatment status",
       x       = "Subject ID ", 
       y       = "Days since hospital admission",
       title   = "Patient trajectories",
       caption ="Hypothetical example")  
  geom_hline(yintercept = 2, linetype = "dashed", size = 1)  
  theme_minimal()  
  theme(plot.title   = element_text(hjust = 0.5),
        plot.caption = element_text(size = 7, hjust = 0))

Data in reproducible format

df <- data.frame(subjectID = factor(1:10, 10:1),
                 stage = rep(c("treated", "untreated"), times = c(6, 4)),
                 startTime = 0,
                 endTime = c(8, 12, 1.7, 1.8, 15,17, 8, 14, 1, 1.5),
                 isContinued = rep(c(!0, !1, !0, !1), times = c(2, 2, 4, 2)),
                 responseType = rep(c("death", "discharge"), 5))

df
#>    subjectID     stage startTime endTime isContinued responseType
#> 1          1   treated         0     8.0        TRUE        death
#> 2          2   treated         0    12.0        TRUE    discharge
#> 3          3   treated         0     1.7       FALSE        death
#> 4          4   treated         0     1.8       FALSE    discharge
#> 5          5   treated         0    15.0        TRUE        death
#> 6          6   treated         0    17.0        TRUE    discharge
#> 7          7 untreated         0     8.0        TRUE        death
#> 8          8 untreated         0    14.0        TRUE    discharge
#> 9          9 untreated         0     1.0       FALSE        death
#> 10        10 untreated         0     1.5       FALSE    discharge