The code you provided creates two plots, one with a color legend for both points and lines (p3) and another plot that is manipulated to include the colors from p1 and p2 as point colors, while keeping the line colors from p2 (pp3). This second approach provides more control over the colors in the legend.

Here’s a brief explanation of how this works:

• The color legends for points and lines are suppressed using theme(legend.position = "none").
• The data from p1 and p2 is extracted, and the corresponding colors are replaced with those from p1 for points (point_cols_vec) and p2 for lines (line_cols_vec).
• A new plot (grob3) is created using ggplot_gtable(pp3), where pp3 combines the manipulated data.
• The plots are arranged side by side with the legends, similar to your original code.

This approach offers more flexibility in controlling the appearance of the color legend. If you want to keep the same colors for points from one plot but different colors for lines from another, this method can be adjusted accordingly.

Example Usage

You can use this approach to create custom plots with specific color schemes or legends:

# Load required libraries
library(ggplot2)
library(gridExtra)

# Create sample data
set.seed(123)
df_points <- data.frame(value = rnorm(100), group = c("A", "B", "C") * 3)
df_lines <- data.frame(x = seq(-1, 1, by = 0.01), y = sin(pi * x))

# Create plot p1 with color legend for points and lines
p1 <- ggplot(df_points, aes(value, group)) +
  geom_point(aes(color = group),
             position = "jitter", alpha = 0.2, size = 2) +
  geom_line() +
  scale_color_manual(values = c("A", "B", "C")) +
  theme(legend.position = "bottom")

p1

# Create plot p2 with color legend for lines
p2 <- ggplot(df_lines, aes(x, y)) +
  geom_line(color = "blue") +
  geom_point(aes(group = group), color = c("red", "green", "black")) +
  scale_color_manual(values = c("A", "B", "C"))

p2

# Create plot p3 with suppressed legend and manipulated colors
pp <- ggplot(df_points, aes(value, group)) +
  geom_point(aes(color = group),
             position = "jitter", alpha = 0.2, size = 2) +
  geom_line(data = df_lines, aes(group = group, color = group), 
            linetype = c("dashed", "solid", "dotted")) +
  scale_color_manual(values = c("A", "B", "C")) +
  theme(legend.position = "none")

pp <- pp + 
  geom_line(data = df_lines, aes(group = group, color = group), 
            linetype = c("dashed", "solid", "dotted"))

pp

# Create a grid of plots with the desired layout
grid.arrange(
  print(p1),
  print(pp),
  ncol = 2,
  widths = c(0.45, 0.55)
)

This example demonstrates how to create three plots: p1 with a suppressed legend for points and lines, pp with manipulated colors, and another plot that combines elements from both approaches.

By adjusting the code according to your needs, you can achieve more complex color schemes or legends in your ggplot2 plots.

Last modified on 2024-10-13