将阴影区域添加到ggplot中手动指定的行

Question

我有Observed和估算（Est）值的回归，如下面的head所示。

data <- structure(list(IndID = structure(c(1L, 2L, 3L, 5L, 6L, 7L, 8L, 
9L, 10L, 11L, 12L, 13L, 14L, 15L, 16L, 17L, 18L, 19L, 20L, 21L, 
22L, 23L), .Label = c("CAL_F01", "CAL_F17", "CAL_F19", "CAL_F23", 
"CAL_F43", "CAL_M33", "CAL_M36", "COL_P01", "COL_P03", "COL_P05", 
"COL_P06", "COL_P07", "COL_P08", "COL_P09", "COL_P10", "COL_P12", 
"COL_P13", "PAT_F03", "PAT_F04", "PAT_F05", "PAT_M02", "PAT_M03", 
"PAT_M04"), class = "factor"), StudyArea = structure(c(1L, 1L, 
1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 3L, 3L, 
3L, 3L, 3L, 3L), .Label = c("Cali", "Colo", "Pata"), class = "factor"), 
    Observed = c(22L, 50L, 8L, 54L, 30L, 11L, 90L, 53L, 9L, 42L, 
    72L, 40L, 60L, 58L, 20L, 37L, 50L, 67L, 20L, 19L, 58L, 5L
    ), variable = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L), .Label = "PredKills", class = "factor"), 
    Est = c(28, 52, 6, 35, 31, 13, 80, 62, 4, 43, 66, 43, 55, 
    42, 20, 47, 36, 84, 20, 17, 36, 6), SE = c(3.50031581162016, 
    4.8742514125436, 1.20589766104628, 4.79430832229519, 3.87541734990744, 
    2.36031827307993, 6.35148447967163, 5.52456747941261, 1.60267125934065, 
    4.53967516735091, 6.61559705260502, 5.35175112687543, 5.89582419295991, 
    5.18042529534246, 3.43767468948519, 4.69809433696684, 3.80733165582324, 
    5.85520173339347, 3.151903629499, 2.64621136787301, 4.64130814363024, 
    1.41537000011436)), .Names = c("IndID", "StudyArea", "Observed", 
"variable", "Est", "SE"), row.names = c(1L, 2L, 3L, 5L, 6L, 7L, 
8L, 9L, 10L, 11L, 12L, 13L, 14L, 15L, 16L, 17L, 18L, 19L, 20L, 
21L, 22L, 23L), class = "data.frame")


> head(data)
    IndID StudyArea Observed  variable Est       SE
1 CAL_F01      Cali       22 PredKills  28 3.500316
2 CAL_F17      Cali       50 PredKills  52 4.874251
3 CAL_F19      Cali        8 PredKills   6 1.205898
5 CAL_F43      Cali       54 PredKills  35 4.794308
6 CAL_M33      Cali       30 PredKills  31 3.875417
7 CAL_M36      Cali       11 PredKills  13 2.360318

此代码如下图所示。

p2 <- ggplot(data, aes(x=Observed, y=Est, color=StudyArea))
p2+ geom_point(shape="*", size = 12) + 
  geom_abline(intercept =0, slope = 1, size = 1)+
  geom_errorbar(aes(x= Observed, ymin=Est-SE, ymax=Est+SE, color=StudyArea),width = 0.5,cex=1, lty=2)+
  scale_color_manual(values=c("red","blue","darkgreen"))+
  ylim(2,92)+ xlim(2,92)

实线表示一对一的拟合。即如果Obs = Est，那么积分就会落在线上。线上的残差显然是误差。

问题：如何将阴影区域添加到代表10％错误的45度线。我认为我需要按照建议at this SO post使用geom_ribbon，但无法生成正确的结果。

ADDITION 阴影区域不应随x轴或y轴的变化而变化，但应在45度线上保持不变。

我想保持x轴值相同，并且通过更改y轴值阴影区域，其中y比Est大0.9并且小于Est。

以下是我一直在使用的代码，用于添加比Est低0.9的额外行。

p2+ geom_point(shape="*", size = 12) + 
  geom_errorbar(aes(x= ObsKills, ymin=value-SE, ymax=value+SE, color=StudyArea),width = 0.5,cex=1, lty=2)+
  coord_cartesian(ylim=c(2,92), xlim=c(2,92))+
  scale_color_manual(values=c("red","blue","darkgreen"))+
  geom_abline(intercept =0, slope = 1, size = 1, col="red")+
  geom_abline(data=data.frame(x=seq(1,92,1),y=seq(1,92,1)), aes(x=x, y=y*0.9),lty=2, cex=1)

使用最后一行代码（从@BrodieG获取提示），我创建了一个新的data.frame，其中seq的值反映了Observed和Est数据。在aes函数中我多y乘以0.9。在我看来，这一行应该比Est低0.9而不是顶部。

我希望在它们之间添加一条较低和较高的线然后遮挡，尽管可能有更好的方法来做到这一点。

希望这更清楚......

提前致谢！

Answer 1

这是一个实现，虽然我不确定我做的正是你想要的：

p2 <- ggplot(data, aes(x=Observed, y=Est, color=StudyArea))
p2+ 
  geom_ribbon(data=data.frame(x=c(0,100)), aes(x=x, ymin=x * .9, ymax=x * 1.1), fill="gray", inherit.aes=F, alpha=0.5) +
  geom_abline(intercept =0, slope = 1, size = 1)+
  geom_point(shape="*", size = 12) + 
  geom_errorbar(aes(x= Observed, ymin=Est-SE, ymax=Est+SE, color=StudyArea),width = 0.5,cex=1, lty=2)+
  scale_color_manual(values=c("red","blue","darkgreen"))+
  coord_cartesian(ylim=c(2,92), xlim=c(2,92))