如何按变量设置ggridges绘图的点形状与填充颜色？

Got it! Since you're using the development version of ggridridges (which adds point display functionality), we can directly map your gender group to point_shape and point_fill to get the exact styling you want—green squares for females, blue triangles for non-females. Here's a step-by-step solution tailored to your needs:

Step 1: Simulate Your Data #

First, let's create a reproducible sample dataset matching your use case:

library(ggplot2)
library(ggridridges)

# Set seed for consistent results
set.seed(123)
df <- data.frame(
  gender = rep(c("Female", "Non-Female"), each = 200),
  metric_value = c(rnorm(200, mean = 5), rnorm(200, mean = 7)),
  category = rep(1:10, 40)
)

Step 2: Build the Ridge Plot with Custom Points #

We’ll map the gender variable to both point_shape and point_fill, then use manual scales to define your preferred styles:

ggplot(df, aes(x = metric_value, y = category, group = gender,
               point_shape = gender, point_fill = gender)) +
  geom_density_ridges(
    jittered_points = TRUE, # Critical for showing points (dev version feature)
    position = position_points_jitter(width = 0.4), # Control point spread
    alpha = 0.6, # Soften ridge transparency
    point_size = 2.5 # Adjust point size for visibility
  ) +
  # Assign shapes: Square (15) for Female, Triangle (17) for Non-Female
  scale_point_shape_manual(values = c("Female" = 15, "Non-Female" = 17)) +
  # Assign fill colors: Green for Female, Blue for Non-Female
  scale_point_fill_manual(values = c("Female" = "#27ae60", "Non-Female" = "#2980b9")) +
  labs(title = "Ridge Plot with Gender-Specific Points",
       x = "Metric Value", y = "Category",
       point_shape = "Gender", point_fill = "Gender") +
  theme_minimal()

Quick Tips: #

• If you haven’t installed the dev version of ggridridges yet, run: devtools::install_github("clauswilke/ggridridges")
• Shapes like 15 (square) and 17 (triangle) are filled shapes—if you wanted unfilled outlines, use shapes like 0 (square) or 2 (triangle) and replace point_fill with point_color.
• Tweak position_points_jitter(width) to adjust how spread out the points appear along the x-axis.

内容的提问来源于stack exchange，提问作者Matti Heino