ggplot2 - 如何将 geom_dotplot 与 x 轴标签对齐？

Question

我使用 ggplot2 在下面生成了点 plot。 然而，这些点与它们各自的实验不一致（第一个实验除外）。 我如何让它们对齐？

这是我使用的脚本：

emibc_plot <- ggplot(emibc, aes(x = experiment, y = value, fill = model)) +
  theme(plot.title = element_text(hjust = 0.5)) +
  theme(legend.title=element_blank()) +
  scale_y_continuous(breaks = sort(c(seq(min(emibc$value), max(emibc$value), length.out=5), 0)), label= function(x) {ifelse(x==0, "0", scales::scientific_format(digits = 3)(x))}) +
  geom_dotplot(binaxis = "y", stackdir = "center", stackgroups=TRUE, binpositions="all", dotsize = 0.5)

这是dput(emibc)的 output ：

structure(list(model = c("CESM", "E3SM", "GISS", "MIROC", "CESM", 
"E3SM", "GISS", "MIROC", "CESM", "E3SM", "GISS", "MIROC", "CESM", 
"E3SM", "MIROC", "CESM", "E3SM", "GISS", "MIROC"), experiment = c("bc_no_seas", 
"bc_no_seas", "bc_no_seas", "bc_no_seas", "high_SO4", "high_SO4", 
"high_SO4", "high_SO4", "no_SO4", "no_SO4", "no_SO4", "no_SO4", 
"SO2_at_height", "SO2_at_height", "SO2_at_height", "SO2_no_season", 
"SO2_no_season", "SO2_no_season", "SO2_no_season"), value = c(2.13359e-16, 
2.13549e-16, 1.04680e-17, 2.86707e-16, 
3.97356e-20, 1.77632e-20, 0, 0, 2.22483e-20, 
1.17037e-20, 0, 0, 2.51571e-20, 9.81121e-21, 
0, 1.67484e-20, 1.46975e-20, 0, 0)), row.names = c(NA, 
-19L), class = c("tbl_df", "tbl", "data.frame"))

可能值得注意的是，我在运行脚本时收到此警告：

`stat_bindot()` using `bins = 30`. Pick better value with `binwidth`.

Answer 1

ggplot2 的 GitHub 存储库中有一些未解决的问题； 见1745和3620 。 这里还有一个拉取请求（已关闭且未合并）： 1096 。

不过，有一种 hacky 方法。 从 PR 1096和这个答案，您可以将fill参数传递给geom_dotplot() 。 正如用户vitorbl 在这里提到的，您的数据似乎必须先排序。

# required libraries
library(ggplot2)
library(dplyr)

# OP's data 
emibc <- structure(list(
  model = c("CESM", "E3SM", "GISS", "MIROC", "CESM", 
            "E3SM", "GISS", "MIROC", "CESM", "E3SM", 
            "GISS", "MIROC", "CESM", "E3SM", "MIROC",
            "CESM", "E3SM", "GISS", "MIROC"), 
  experiment = c("bc_no_seas", "bc_no_seas", "bc_no_seas",
                 "bc_no_seas", "high_SO4", "high_SO4", 
                 "high_SO4", "high_SO4", "no_SO4", 
                 "no_SO4", "no_SO4", "no_SO4", "SO2_at_height",
                 "SO2_at_height", "SO2_at_height", "SO2_no_season", 
                 "SO2_no_season", "SO2_no_season", "SO2_no_season"), 
  value = c(2.13359e-16, 2.13549e-16, 1.04680e-17, 
            2.86707e-16, 3.97356e-20, 1.77632e-20,
            0, 0, 2.22483e-20, 1.17037e-20, 0, 0,
            2.51571e-20, 9.81121e-21, 0, 
            1.67484e-20, 1.46975e-20, 0, 0)), 
  row.names = c(NA, -19L), 
  class = c("tbl_df", "tbl", "data.frame"))

# order data
emibc <- emibc %>% 
  arrange(experiment, value)

# create a column with a color for each experiment
emibc <- emibc %>% 
  mutate(
    color =  case_when(
      model == "CESM" ~ "red",
      model == "E3SM" ~ "darkgreen",
      model == "GISS" ~ "lightblue",
      model == "MIROC" ~ "purple"
    )
  )

# make plot
ggplot(emibc, aes(x = experiment, y = value)) +
  theme(plot.title = element_text(hjust = 0.5)) +
  theme(legend.title = element_blank()) +
  scale_y_continuous(breaks = sort(c(seq(min(emibc$value), max(emibc$value), 
                                         length.out=5), 0)), 
                     label = function(x) {ifelse(x==0, "0", scales::scientific_format(digits = 3)(x))}) +
  geom_dotplot(binaxis = "y", stackdir = "center", dotsize = 0.5, 
               fill = emibc$color)