r - R中带有分组列的ggplot2条形图，如何取消堆叠？

Question

我正在尝试将数据(见下文)绘制为分组条形图。我已按照此处的其他建议进行更改 position = ""在 geom_bar ，但我似乎无法让它发挥作用。我得到的只是一个堆积的条形图。这是我最近的尝试:

ggplot(data = IFNg, aes(x = IFNg$Location, y = IFNg$`Fold Change`)) + 
    geom_bar(position = 'identity', stat = "identity", fill = IFNg$Day) + 
    ylab("Fold Change (log_2)") + geom_errorbar(limits, width=0.15)

我还尝试了以下方法，这也会导致堆积图，但这个图的颜色不同:

ggplot(data = IFNg, aes(x = IFNg$Location, y = IFNg$`Fold Change`, 
    fill= IFNg$Day)) + geom_bar(position = 'identity', stat = "identity") + 
    ylab("Fold Change (log_2)") + geom_errorbar(limits, width=0.15)

这是我的其余代码与数据:

dat <- structure(list(Day = c(1L, 3L, 7L, 21L, 1L, 3L, 7L, 21L, 1L, 
3L, 7L, 21L), Order = c(1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 
3L, 3L), Target = c("IFN-g", "IFN-g", "IFN-g", "IFN-g", "IFN-g", 
"IFN-g", "IFN-g", "IFN-g", "IFN-g", "IFN-g", "IFN-g", "IFN-g"), 
Location = c("Duodenum", "Duodenum", "Duodenum", "Duodenum", 
"Duodenum (Lymph)", "Duodenum (Lymph)", "Duodenum (Lymph)", 
"Duodenum (Lymph)", "Proximal Jejunum", "Proximal Jejunum",
"Proximal Jejunum", "Proximal Jejunum"), `Fold Change` = c(-1.750896098,
-2.194886907, -0.838822724, 0.680612172, -0.375563984, 0.650154987, 
0.844897327, -0.228441603, -1.298230671, -1.173634963, -1.877736135, 
0.787322978), StDev = c(0.027477611, 0.029974897, 0.19025044, 1.235986647,
0.055798435, 0.086102327, 0.115422155, 0.34470734, 0.020947691, 0.165294027,
0.040111751, 0.035010207)), .Names = c("Day", "Order", "Target", "Location",
"Fold Change", "StDev"), class = c("tbl_df", "data.frame"), row.names = c(NA,-12L))

IFNg <- dat[dat$Target == 'IFN-g', ]

limits <- aes(ymax = IFNg$`Fold Change` + IFNg$StDev,
    ymin = IFNg$`Fold Change` - IFNg$StDev)

Answer 1

您的大部分问题都来自您输入了不应该输入的内容，例如 data$column而不是 column .另外，您的 Day是数字，但具有由 Day 定义的不同并排条，它需要是离散的(一个因素)。

我还要指出的是?geom_bar帮助页面并不长，有这一段:

By default, multiple x's occurring in the same place will be stacked atop one another by position_stack. If you want them to be dodged side-to-side, see ?position_dodge. Finally, position_fill shows relative proportions at each x by stacking the bars and then stretching or squashing to the same height.

通常， position = 'dodge'会工作得很好，但它会感到困惑，并认为在这种情况下您的误差条宽度是要躲避的宽度，因此我们将使用完整的 position = position_dodge() :

# No data$ inside aes()!!!
limits <- aes(ymax = `Fold Change` + StDev,
    ymin = `Fold Change` - StDev)

ggplot(data = IFNg,
  # group lets `ggplot` know we want different errorbars/bars for each day
       aes(x = Location, y = `Fold Change`, group = factor(Day))
    ) +
    geom_bar(
        stat = "identity",
        aes(fill = factor(Day)),
        position = position_dodge(width = 0.9)
    ) +
    ylab("Fold Change (log_2)") +
    geom_errorbar(
        limits,
        width = 0.15,
        position = position_dodge(width = 0.9)
    )