将变量汇总的点与由同一变量躲避的点对齐

Question

我有来自两个不同数据集的数据。第一个，dat1包含多个点。第二个dat2仅包含dat1中每个Season-Species组的最大值。我试图绘制dat1，然后想要绘制更大的形状，突出显示每个Season-Species组的最大值，即dat2。

数据：

library(ggplot2)
library(dplyr)

dat1 <- structure(list(Season = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 2L, 
    2L, 2L, 2L, 2L, 2L), .Label = c("Summer", "Winter"), class = "factor"), 
        Species = structure(c(1L, 1L, 1L, 2L, 2L, 2L, 1L, 1L, 1L, 
        2L, 2L, 2L), .Label = c("BHS", "MTG"), class = "factor"), 
        CovGrain = structure(c(1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L, 
        1L, 2L, 3L), .Label = c("CanCov_30", "CanCov_500", "CanCov_1000", 
        "NDVI_30", "NDVI_500", "NDVI_1000", "Slope_30", "Slope_500", 
        "Slope_1000", "SlopeVar_30", "SlopeVar_500", "SlopeVar_1000"
        ), class = "factor"), Count = c(4L, 19L, 4L, 5L, 3L, 14L, 
        14L, 9L, 9L, 4L, 10L, 9L)), .Names = c("Season", "Species", 
    "CovGrain", "Count"), class = "data.frame", row.names = c(1L, 
    2L, 3L, 14L, 15L, 16L, 30L, 31L, 32L, 45L, 46L, 47L))

dat2 <- dat1 %>% group_by(Season, Species) %>% 
  filter(Count == max(Count)) %>% as.data.frame()


ggplot(dat1, aes(x = CovGrain, y = Count)) +
  geom_point(aes(fill = Species, color = Species),
             alpha = 0.5, stroke = 3, size = 3, position=position_dodge(0.5)) +
  facet_wrap(~Season, scales = "free_x") +
  scale_shape_manual(values = c(21,22,24)) +
  scale_fill_manual(values=c("blue", "red")) +
  geom_point(data = dat2, aes(x = CovGrain, y = Count), shape = 23, 
             stroke = 2, size = 6, position=position_dodge(0.5)) +
  theme_bw()

在下图中，我想要正确地避开黑色三角形，以便勾勒出每组的最大点。

任何建议都表示赞赏！

Answer 1

这有点hacky，并且可能有更好的解决方案，但一种方法是使用您自己的随机性创建一个新的x变量。 hacky部分来自于首先执行geom_point(size = -1)以使其保持x轴。所以，不管怎么说，不是优雅的，但是我想你会得到你想要的东西。

dat1$id <- 1:nrow(dat1)

dat2 <- dat1 %>% 
    group_by(Season, Species, Cov) %>% 
    filter(Count == max(Count)) %>% 
    as.data.frame()

randomness <- rnorm(nrow(dat1), 0, 0.5)
dat1$new_x <- as.integer(dat1$CovGrain) + randomness
dat2$new_x <- as.integer(dat2$CovGrain) + randomness[dat2$id]

ggplot(dat1, aes(x = CovGrain, y = Count)) +
  geom_point(size = -1) +
  geom_point(aes(x = new_x, fill = Species, color = Species),
             alpha = 0.5, stroke = 3, size = 3) +
  facet_wrap(~Season, scales = "free_x") +
  scale_shape_manual(values = c(21,22,24)) +
  scale_fill_manual(values=c("blue", "red")) +
  geom_point(data = dat2, aes(x = new_x, y = Count), shape = 23, 
             stroke = 2, size = 6) +
  theme_bw()

Answer 2

在原始数据中创建一个布尔变量，指示'Count'是否最大，按'Season'和'Species'分组。使用scale_alpha_manual将alpha设置为0 FALSE（即“计数”不在max）。使用group = Species以“物种”躲闪。

dat1 <- dat1 %>% group_by(Season, Species) %>%
  mutate(max_count = Count == max(Count))

pos <- position_dodge(0.5)

ggplot(dat1, aes(x = CovGrain, y = Count)) +
  geom_point(aes(color = Species), position = pos) +
  geom_point(aes(alpha = max_count, group = Species), shape = 23, size = 6, position = pos) +
  facet_wrap(~ Season) +
  scale_alpha_manual(values = c(0, 1), guide = "none") +
  theme_bw()