如何让ggplot2对数转换后的X轴从0开始并正确缩放？

使用ggplot2绘制Gapminder数据集的人均GDP（gdpPercap）与预期寿命（lifeExp）散点图，要求经过对数转换的X轴从0开始并正确缩放。当前代码如下：

主绘图代码 #

xbreaks <- c(0, 10000, 20000, 30000, 40000, 50000)
popbreaks <- rev(c(3, 10, 30, 100, 300)*1000000)

### Golden Scatter attempt
p1 <- ggplot(data = mainData,
            mapping = aes(x= gdpPercap,
                          y = lifeExp,
                          color = continent))

p <- p1 + goldenScatterCAtheme+  
      geom_point(alpha = 0.55, aes(size=pop)) +
      scale_x_log10(breaks= xbreaks,
                    labels= xbreaks/1000) +
      scale_color_brewer(palette = "Set1") +
      scale_size(breaks = popbreaks) +
      labs(x = "GDP per Capita", y = "Life Expectancy in Years")
      
p <- p + guides(size = FALSE)
#p <- p + scale_x_continuous(expand = c(0, 0), limits = c(0,NA))
      
width <- 7
ggsave("ggScatterHw1.pdf", width = width, height = width/1.618)

自定义主题代码 #

goldenScatterCAtheme <- theme(
  panel.background = element_rect(fill = "white"),
  aspect.ratio = ((1 + sqrt(5))/2)^(-1),
  axis.ticks.length = unit(0.5, "char"),
  #Horizontal
  axis.line.x.top = element_line(size = 0.1),
  axis.line.x.bottom = element_line(size = 0.1),
  axis.ticks.x = element_line(size = 0.1),
  axis.text.x = element_text(color = "gray45", size = 12),
  axis.title.x = element_text(size = 12, margin = margin(t = 7.5, r = 0, b = 0, l = 0)),
  #Vertical
  axis.ticks.y = element_line(size = 0.1),
  axis.text.y = element_text(color = "gray45", size = 12, margin = margin(t = 0, r = 0, b = 0, l = 0)),
  axis.title.y = element_text(size = 12, margin = margin(t = 0, r = 7.5, b = 0, l = 0)),
  #Legend
  legend.key = element_rect(fill = NA, color = NA),
  #GridLines
  panel.grid.major.x = element_line(color = "gray95", size = 0.2),
  panel.grid.major.y = element_line(color = "gray95", size = 0.2),
  #Faceting
  strip.background = element_blank(),
  strip.text.x = element_text(size = 12),
  strip.text.y = element_blank(),
  strip.placement = "outside",
  panel.spacing.x = unit(1.25, "lines"),
  panel.spacing.y = unit(1, "lines")
)
class(goldenScatterCAtheme)

当前图表效果：

对数函数在x=0处无定义，直接用scale_x_log10无法将轴起点设为0，可通过以下两种方法实现需求：

方法1：伪对数转换（推荐） #

伪对数转换可处理0值，同时对大数保持对数缩放特性，适合让轴从0开始且保留对数缩放的场景：

scale_x_continuous(
  trans = scales::pseudo_log_trans(sigma = 100),  # sigma控制0附近线性区域宽度，可按需调整
  breaks = xbreaks,
  labels = xbreaks/1000,
  expand = c(0, 0),
  limits = c(0, NA)
)

• sigma参数：数值越小，0附近线性区域越窄；针对Gapminder的gdpPercap数据，设为100能让缩放更自然。
• expand = c(0,0)和limits = c(0,NA)确保轴起点对齐0。

方法2：视觉模拟0起点（保留纯对数转换） #

若坚持使用纯对数转换，可通过调整轴扩展和断点标签，视觉上模拟轴从0开始：

scale_x_log10(
  breaks = xbreaks[-1],  # 移除无效的0断点
  labels = c("0", xbreaks[-1]/1000),  # 手动添加0的标签
  expand = expansion(mult = c(0.05, 0)),  # 向左扩展轴范围，靠近0
  limits = c(min(mainData$gdpPercap)/2, NA)  # 设下限为数据最小值的一半，避免截断点
)

此方法仅视觉上接近0，实际对数轴仍从正数开始，适合严格需要纯对数缩放的场景。

完整修改代码（方法1） #

xbreaks <- c(0, 10000, 20000, 30000, 40000, 50000)
popbreaks <- rev(c(3, 10, 30, 100, 300)*1000000)

p1 <- ggplot(data = mainData,
            mapping = aes(x= gdpPercap,
                          y = lifeExp,
                          color = continent))

p <- p1 + goldenScatterCAtheme+  
      geom_point(alpha = 0.55, aes(size=pop)) +
      scale_x_continuous(
        trans = scales::pseudo_log_trans(sigma = 100),
        breaks = xbreaks,
        labels = xbreaks/1000,
        expand = c(0, 0),
        limits = c(0, NA)
      ) +
      scale_color_brewer(palette = "Set1") +
      scale_size(breaks = popbreaks) +
      labs(x = "GDP per Capita", y = "Life Expectancy in Years")
      
p <- p + guides(size = FALSE)
      
width <- 7
ggsave("ggScatterHw1.pdf", width = width, height = width/1.618)

内容的提问来源于stack exchange，提问作者Kenzie DeKeyser