Question

我用来自不同数据帧的多个geom对象制作了一个ggplot。我有一个模型输出，该输出为数据框格式，具有深度（y轴值）列，2.5％，50％和97.5％的置信区间范围。此处缩短了数据集：

head(model_output)
     Depth     2.5%    50%    97.5%
1 32.00000  767.000  888.5  977.025
2 33.30303 1085.925 1207.0 1303.025
3 34.60606 1356.000 1431.0 1522.000
4 35.90909 1362.975 1444.0 1531.000
5 37.21212 1369.975 1453.0 1548.000
6 38.51515 1376.000 1460.5 1563.025

我还具有许多样本的概率密度函数，这里显示了三个。

head(sample1)
  positions ageGrid    densities
1        32     686 1.175521e-05
2        32     687 1.543893e-05
3        32     688 1.812884e-05
4        32     689 1.963192e-05
5        32     690 2.125031e-05
6        32     691 2.125031e-05

head(sample2)
  positions ageGrid    densities
1        75    1538 1.084429e-05
2        75    1539 1.617676e-05
3        75    1540 1.963299e-05
4        75    1541 2.681821e-05
5        75    1542 3.031380e-05
6        75    1543 3.534482e-05

head(sample3)
  positions ageGrid    densities
1        75    2349 1.204547e-05
2        75    2350 1.673264e-05
3        75    2351 2.306552e-05
4        75    2352 3.154642e-05
5        75    2353 4.280118e-05
6        75    2354 5.759857e-05

我可以将所有这些与下面的代码一起绘制。置信区间为geom_ribbon和geom_line对象，三个样本pdf为geom_ridgeline对象。

ggplot(data = model_output, aes(x = `50%`, y = Depth))+
  geom_ribbon(data = model_output, aes(xmin = `2.5%`, xmax = `97.5%`), fill = "grey50", colour = "grey50", alpha = 0.5)+
  geom_line(data = model_output, aes(x = `50%`, y = Depth), colour = "black", size = 1)+
  scale_y_reverse()+
  scale_x_reverse()+
  theme_bw()+
  xlab("x axis")+
  ylab("y axis")+
  labs(title = "title", subtitle = "subtitle")+
  geom_ridgeline(data = sample1,
                 aes(x = ageGrid, y = positions, height = densities*1000, group = 'sample1'),
                 fill = "steelblue", colour = "steelblue", alpha = 0.6)+
  geom_ridgeline(data = sample2,
                 aes(x = ageGrid, y = positions, height = densities*1000, group = 'sample2'),
                 fill = "darkorange2", colour = "darkorange2", alpha = 0.6)+
  geom_ridgeline(data = sample3,
                 aes(x = ageGrid, y = positions, height = densities*1000, group = 'sample3'),
                 fill = "steelblue", colour = "steelblue", alpha = 0.6)

我发现这个图很好，正是我想要的图。但是我也想要一个图例，它显示了我在此处应用的几何图形和填充颜色：

作为灰色geom_ribbon对象的置信区间
作为黑色geom_line对象的置信区间中值
以及带有geom_ridgeline对象的两种不同颜色的样本（因为有两种类型的样本）（一种蓝色和一种橙色）。

我已经尝试了以下问题/解答How to add legend to ggplot manually? - R，Missing legend with ggplot2 and geom_line，Adding manual legend in ggplot，Adding manual legend to a ggplot，并提出了以下代码，但给出以下代码无效我想要的传说...

ggplot(data = model_output, aes(x = `50%`, y = Depth))+
  geom_ribbon(data = model_output, aes(xmin = `2.5%`, xmax = `97.5%`, fill = "grey50", colour = "grey50"), alpha = 0.5)+
  geom_line(data = model_output, aes(x = `50%`, y = Depth, colour = "black"), size = 1)+
  scale_y_reverse()+
  scale_x_reverse()+
  theme_bw()+
  xlab("x axis")+
  ylab("y axis")+
  labs(title = "title", subtitle = "subtitle")+
  scale_colour_manual(values = c("grey50", "black", "steelblue", "darkorange"),
                      labels = c("95% CI", "median (50% quartile)", "sample_type1", "sample_type2"))+
  geom_ridgeline(data = sample1,
                 aes(x = ageGrid, y = positions, height = densities*1000, group = 'sample1', 
                     fill = "steelblue", colour = "steelblue", alpha = 0.6),
                 show.legend = TRUE)+
  geom_ridgeline(data = sample2,
                 aes(x = ageGrid, y = positions, height = densities*1000, group = 'sample2',
                     fill = "darkorange2", colour = "darkorange2", alpha = 0.6), show.legend = TRUE)+
  geom_ridgeline(data = sample3,
                 aes(x = ageGrid, y = positions, height = densities*1000, group = 'sample3',
                     fill = "steelblue", colour = "steelblue", alpha = 0.6))

我是否在错误的位置使用彩色对象（是否在es参数中？）？我尝试融合数据集，但是对于像我这样的初学者来说有点复杂...

ggplot中的多个geom对象的手动图例

0 个答案: