R ggplot2：在多个变量的相同图中添加均值和标准差

Question

我有4个向量，每个向量16个值;每个值都是一个项目的平均值，我在4个数据集中有相同的16个项目。

我使用ggplot2绘制这些意思：这里是一个可重复的例子。

library("ggplot2")
library("dplyr")    

means <- as.data.frame(cbind(rnorm(16),rnorm(16), rnorm(16), rnorm(16)))
means <- mutate(means, id = rownames(means))
colnames(means)<-c("1", "2", "3", "4", "Symptoms")
means_long <- melt(means, id="Symptoms")
means_long$Symptoms <- as.numeric(means_long$Symptoms)
names(means_long)[2] <- "Datasets"

ggplot(data=means_long, aes(x=Symptoms, y=value, colour=Datasets)) +
      geom_line() +
      geom_point(shape = 21, fill = "white", size = 1.5, stroke = 1) +
      xlab("Symptoms") + ylab("Means") +
      scale_y_continuous() + 
      scale_x_continuous(breaks=c(1:16)) +
      theme_bw() +
      theme(panel.grid.minor=element_blank()) +
      coord_flip()

现在，我有4个其他向量，这是4个数据集的16个项目的标准偏差。我想将它们绘制成相同的情节。数据格式与上面相同，因此它实际上是相同的代码。

我希望同一图中的标准偏差与平均值相同，使用相同颜色但不同的线类型（因此数据集1均值为红色，数据集1标准差为虚线），最好的情况下是区分两者的图例除了线条和虚线的“平均值”与“标准偏差”之外，还有数据集（我目前）。

感谢您的帮助！

Answer 1

这有帮助吗？

为了使它看起来不那么丑陋，我将所有随机平均值都设为正值，然后将示例标准偏差设为负值。在同一图表上绘制值的方法是将数据集分别输入到每个geom，而不是在初始ggplot()函数中定义。

如果这不是您的想法，请告诉我：

library("ggplot2")
library("dplyr")    

means <- as.data.frame(abs(cbind(rnorm(16),rnorm(16), rnorm(16), rnorm(16))))
means <- mutate(means, id = rownames(means))
colnames(means)<-c("1", "2", "3", "4", "Symptoms")
means_long <- reshape2::melt(means, id="Symptoms")
means_long$Symptoms <- as.numeric(means_long$Symptoms)
names(means_long)[2] <- "Datasets"


sds_long <- means_long
sds_long$value <- -sds_long$value

ggplot() +
  geom_line(aes(x=Symptoms, y=value, colour=Datasets), lty=1, data=means_long) +
  geom_point(aes(x=Symptoms, y=value, colour=Datasets), data=means_long, shape = 21, fill = "white", size = 1.5, stroke = 1) +
  geom_line(aes(x=Symptoms, y=value, colour=Datasets), lty=2, data=sds_long) +
  geom_point(  aes(x=Symptoms, y=value, colour=Datasets), data=sds_long, shape = 21, fill = "white", size = 1.5, stroke = 1) +
  xlab("Symptoms") + ylab("Means") +
  scale_y_continuous() + 
  scale_x_continuous(breaks=c(1:16)) +
  theme_bw() +
  theme(panel.grid.minor=element_blank()) +
  coord_flip()

＃

回答您的图例查询。简而言之，我认为这很难，因为两个数据集都使用相同的映射美学。

然而，使用code from this answer - 我做了以下事情。我们的想法是从两个情节中获取图例，只绘制平均值/ sds，然后将这些图例添加到没有图例的情节版本中。它可以进行调整，以便您更加手动定位图例...

### Step 1
# Draw a plot with the colour legend
p1 <- ggplot() +
  geom_line(aes(x=Symptoms, y=value, colour=Datasets), lty=1, data=means_long) +
  geom_point(aes(x=Symptoms, y=value, colour=Datasets), data=means_long, shape = 21, fill = "white", size = 1.5, stroke = 1) +
  scale_color_manual(name = "Means",values=c("red","blue", "green","pink")) +
  coord_flip()+
  theme_bw() +
  theme(panel.grid.minor=element_blank()) +
  theme(legend.position = "top")

# Extract the colour legend - leg1
library(gtable)
leg1 <- gtable_filter(ggplot_gtable(ggplot_build(p1)), "guide-box") 

### Step 2
# Draw a plot with the size legend
p2 <- ggplot() +
    geom_line(aes(x=Symptoms, y=value, color=Datasets), lty=2, data=sds_long) +
    geom_point(  aes(x=Symptoms, y=value, color=Datasets), data=sds_long, shape = 21, fill = "white", size = 1.5, stroke = 1) +
    coord_flip()+
  theme_bw() +
  theme(panel.grid.minor=element_blank()) +
  scale_color_manual(name = "SDs",values=c("red","blue", "green","pink")) 

# Extract the size legend - leg2
leg2 <- gtable_filter(ggplot_gtable(ggplot_build(p2)), "guide-box") 

# Step 3
# Draw a plot with no legends - plot
p3<-ggplot() +
  geom_line(aes(x=Symptoms, y=value, colour=Datasets), lty=1, data=means_long) +
  geom_point(aes(x=Symptoms, y=value, colour=Datasets), data=means_long, shape = 21, fill = "white", size = 1.5, stroke = 1) +
  geom_line(aes(x=Symptoms, y=value, color=Datasets), lty=2, data=sds_long) +
  geom_point(  aes(x=Symptoms, y=value, color=Datasets), data=sds_long, shape = 21, fill = "white", size = 1.5, stroke = 1) +
  xlab("Symptoms") + ylab("Means") +
  scale_y_continuous() + 
  scale_x_continuous(breaks=c(1:16)) +
  theme_bw() +
  theme(panel.grid.minor=element_blank()) +
  coord_flip()+
  scale_color_manual(values=c("red","blue", "green","pink")) +
  theme(legend.position = "none")

### Step 4
# Arrange the three components (plot, leg1, leg2)
# The two legends are positioned outside the plot: 
# one at the top and the other to the side.
library(grid)
plotNew <- arrangeGrob(leg1, p3, 
                       heights = unit.c(leg1$height, unit(1, "npc") - leg1$height), ncol = 1)

plotNew <- arrangeGrob(plotNew, leg2,
                       widths = unit.c(unit(1, "npc") - leg2$width, leg2$width), nrow = 1)

grid.newpage()
grid.draw(plotNew)

Answer 2

我建议您只需要一个数据帧来绘制一个实现。另外，您不需要对代码进行太多调整，但您仍然可以区分数据集（即1,2,3,4）和值类型（例如，均值，sd）。

library("ggplot2")
library("dplyr")    

# Means
means <- as.data.frame(cbind(rnorm(16),rnorm(16), rnorm(16), rnorm(16)))
means <- mutate(means, id = rownames(means))
colnames(means)<-c("1", "2", "3", "4", "Symptoms")
means_long <- melt(means, id="Symptoms")
means_long$Symptoms <- as.numeric(means_long$Symptoms)
names(means_long)[2] <- "Datasets"

# Sd
sds_long <- means_long
sds_long$value <- -sds_long$value

################################################################################
# Add "Type" column to distinguish means and sds
################################################################################
type <- c("Mean")
means_long <- cbind(means_long, type)

type <- c("Sd")
sds_long <- cbind(sds_long, type)

merged <- rbind(means_long, sds_long)

colnames(merged)[4] <- "Type"

################################################################################
# Plot
################################################################################
ggplot(data = merged) +
  geom_line(aes(x = Symptoms, y = value, col = Datasets, linetype = Type)) +
  geom_point(aes(x = Symptoms, y = value, col = Datasets), 
             shape = 21, fill = "white", size = 1.5, stroke = 1) +
  xlab("Symptoms") + ylab("Means") +
  scale_y_continuous() + 
  scale_x_continuous(breaks=c(1:16)) +
  theme_bw() +
  theme(panel.grid.minor=element_blank()) +
  coord_flip()