主页 / coding / 问题 / 79529355
Accepted
Mohan
Asked: 2025-03-24 01:26:36 +0800 CST

ggplot：如何将两个图形并排放置并扩展画布以保持默认布局？

  • 772

我有一些 ggplot，我对其进行了调整以确保注释完全符合图的界限。例如：

在此处输入图片描述

现在我想将其中两幅图像并排放置。我尝试使用 grid.arrange 和 patchwork 来实现这一点，但图形的布局发生了变化，大概是因为它被放入了更小的空间：

在此处输入图片描述

一切都变得更加拥挤，点都挤在一起，因此更难阅读，并且注释框偏离了图表的一侧。

我想要做的是将画布扩展至刚好足够大，以使每个图形看起来与单独绘制时完全一样（即使用默认画布大小绘制时）。它可能需要比原来的两倍多一点，以便留出填充空间。我该如何实现这一点？

示例代码：

library(ggplot2)
library(ggtext)
library(tidyverse)
library(patchwork)

N <- 10000
rho = 0.5
  
mu <- c(0, 0)
sigma <- matrix(c(1,   rho,
                  rho, 1   ), 
                nrow = 2, ncol = 2, byrow = TRUE)
            
vs = MASS::mvrnorm(N, mu, sigma)
df = tibble(rMZall = vs[,1], rDZall = vs[,2], n_twin_pairs = runif(N, 5000, 50000))

plot <- function(min_twin_pairs) {
  ggplot(data=df |> filter(n_twin_pairs > min_twin_pairs), aes(rMZall, rDZall)) + 
    ggtitle(sprintf("Studies with > %d twin pairs", min_twin_pairs)) + 
    theme(legend.position="none") +
    # Bounds
    scale_x_continuous(name = "rMZ", limits = c(-0.05, 1.1), breaks = seq(0,1,0.25)) +
    scale_y_continuous(name = "rDZ", limits = c(-0.05, 1), breaks = seq(0,1,0.25)) +
    coord_fixed() +
    # Scatterplot
    stat_density2d(aes(fill=..level.., alpha=..level..), geom='polygon', colour='black', linewidth = 0.1) + 
    scale_fill_continuous(low="green", high="red") +
    geom_point(aes(size = sqrt(n_twin_pairs))) +
    scale_size_continuous(range = c(0, 1)) + 
    # AE and AD lines
    geom_abline(intercept = 0, slope = .5) +
    annotate(geom = "richtext", x = 1, y = .5, label = "AE model", angle = 26.6) + 
    geom_abline(intercept = 0, slope = .25) +
    annotate(geom = "richtext", x = 1, y = .25, label = "AD model", angle = 14.0) 
}

plot1000 = plot(min_twin_pairs = 1000)
plot4000 = plot(min_twin_pairs = 4000)

plot1000
plot4000

plot1000 + plot4000

1 个回答

  1. Best Answer

    您可以改用geomtextpath，特别是geomtextpath::geom_labelabline()。它会自动处理定位，因此您无需摆弄角度、画布大小等。

    library(tidyverse)
library(geomtextpath)
library(patchwork)
set.seed(13)

plot <- function(min_twin_pairs) {
  ggplot(data=df |> filter(n_twin_pairs > min_twin_pairs), aes(rMZall, rDZall)) + 
    ggtitle(sprintf("Studies with > %d twin pairs", min_twin_pairs)) + 
    theme(legend.position="none") +
    # Bounds
    scale_x_continuous(name = "rMZ", limits = c(-0.05, 1.1), breaks = seq(0,1,0.25)) +
    scale_y_continuous(name = "rDZ", limits = c(-0.05, 1), breaks = seq(0,1,0.25)) +
    coord_fixed() +
    # Scatterplot
    stat_density2d(aes(fill=..level.., alpha=..level..), geom='polygon', colour='black', linewidth = 0.1) + 
    scale_fill_continuous(low="green", high="red") +
    geom_point(aes(size = sqrt(n_twin_pairs))) +
    scale_size_continuous(range = c(0, 1)) + 
    # AE and AD lines
    geom_labelabline(intercept = 0, slope = .5, label = "AE model", hjust = 0.85) +
    geom_labelabline(intercept = 0, slope = .25, label = "AD model", hjust = 0.85)
}

plot1000 <- plot(min_twin_pairs = 1000)
plot4000 <- plot(min_twin_pairs = 4000)

plot1000 + plot4000

    • 3

相关问题