5、科研作图之条形图详解——ggplot2

1、简单条形图

数据背景：测定3种浓度下某物质的含量。这里浓度仍然可以看做factor型的数据，数据类型和折线图的数据一样。

基础语法：

library(ggplot2)

df = data.frame(
  c = c(0.5,1,1.5),
  m = c(11.2,13.5,16.1)
)

b = ggplot(df, aes(x=c, y=m))
b + geom_bar(fill = "lightblue", stat = "identity")

添加数值：

使用geom_text()添加文字，主要目的计算y值。

b + geom_bar(fill = "lightblue", stat = "identity")+
  geom_text(aes(y = m - 0.2*m, label = m), vjust = 1.6,
            color = "black", size = 3.5) 

添加误差棒：

同样要先处理数据，计算sd和mean。

library("dplyr")

df = data.frame(
  c = c(0.5,0.5,0.5,1,1,1,1.5,1.5,1.5),
  m = c(11.2,11,12,13,12.1,13.5,15.2,15,16.1)
)
df$c = factor(df$c)
df2 <- df %>%
  group_by(c) %>%
  summarise(
    sd = sd(m),
    m = mean(m)
  )
df2
b = ggplot(df2, aes(x=c, y=m))
b = b + geom_bar(fill = "lightblue", color = "black",stat = "identity")
b + geom_errorbar(aes(x = c, y = m,
                      ymin = m-sd, ymax = m+sd,
                      width = 0.2
))

 

2、多组数据条形图

这里的数据相当于有2种物质在3种浓度下的含量。

基本语法：

型1：

df = data.frame(
  t = c(1,1,1,1,1,1,1,1,1,
        2,2,2,2,2,2,2,2,2),
  c = c(0.5,0.5,0.5,1,1,1,1.5,1.5,1.5,
        0.5,0.5,0.5,1,1,1,1.5,1.5,1.5),
  m = c(11.2,11,12,13,12.1,13.5,15.2,15,16.1,
        10.2,10,11,12,12.1,11.5,14.2,16,16.1)
)
df$c = factor(df$c)
df$t = factor(df$t)

df2 <- df %>%
  group_by(c,t) %>%
  summarise(
    sd = sd(m),
    m = mean(m)
  )

b = ggplot(df2, aes(x=c, y=m))
b = b + geom_bar(aes(fill = t), position = "dodge",stat = "identity")
b

 添加误差线：

b + geom_errorbar(aes(color = t,ymin = m-sd, ymax = m+sd),
                   position ="dodge"
                   )

 型2：

b1 = ggplot(df2, aes(x=c, y=m))
b1 = b1+ geom_bar(aes(fill = t),stat = "identity")
b1