Break Plot

Example

Example of a combo chart with breakpoints.

Setup

• System Requirements: Cross-platform (Linux/MacOS/Windows)

• Programming language: R

• Dependent packages: ggplot2; ggbreak; dplyr; ggpubr; RColorBrewer; rstatix

# Installing necessary packages
if (!requireNamespace("ggplot2", quietly = TRUE)) {
  install.packages("ggplot2")
}
if (!requireNamespace("ggbreak", quietly = TRUE)) {
  install.packages("ggbreak")
}
if (!requireNamespace("dplyr", quietly = TRUE)) {
  install.packages("dplyr")
}
if (!requireNamespace("ggpubr", quietly = TRUE)) {
  install.packages("ggpubr")
}
if (!requireNamespace("RColorBrewer", quietly = TRUE)) {
  install.packages("RColorBrewer")
}
if (!requireNamespace("rstatix", quietly = TRUE)) {
  install.packages("rstatix")
}

# Load packages
library(ggplot2)
library(ggbreak)
library(dplyr)
library(ggpubr)
library(RColorBrewer)
library(rstatix)

Data Preparation

# Data Preparation
df <- ToothGrowth %>%
  group_by(supp, dose) %>%
  summarise(
    mean_len = mean(len),
    sd_len = sd(len),
    n = n(),
    se_len = sd_len/sqrt(n),
    .groups = 'drop')


# Statistical tests (key repair points)
stat.test <- ToothGrowth %>%
  group_by(dose) %>%
  t_test(len ~ supp) %>%
  add_xy_position(x = "dose", dodge = 0.8)

head(df)

# A tibble: 6 × 6
  supp   dose mean_len sd_len     n se_len
  <fct> <dbl>    <dbl>  <dbl> <int>  <dbl>
1 OJ      0.5    13.2    4.46    10  1.41 
2 OJ      1      22.7    3.91    10  1.24 
3 OJ      2      26.1    2.66    10  0.840
4 VC      0.5     7.98   2.75    10  0.869
5 VC      1      16.8    2.52    10  0.795
6 VC      2      26.1    4.80    10  1.52 

Visualization

1. Basic BarPlot

Use basic functions to draw the caption and description of the image.

# Basic BarPlot
p1 <- ggplot(df, aes(x=dose, y=mean_len, fill=supp)) +
  geom_col(position=position_dodge(0.4), width=0.2) +
  geom_errorbar(aes(ymin=mean_len-sd_len, ymax=mean_len+sd_len),
                width=0.1, position=position_dodge(0.4)) +
  scale_y_continuous(breaks = seq(0, 30, 5)) +
  scale_y_cut(breaks=c(15), which=1, scales=1.5) +
  labs(x="Dose (mg/day)", y="Tooth Length (mm)") +
  theme_classic()

p1

Figure 1: Basic BarPlot

2. Grouped scatter plot

# Grouped scatter plot
p2 <- ggplot(ToothGrowth, aes(x=dose, y=len, color=supp)) +
  geom_point(position=position_jitterdodge(jitter.width=0.2)) +
  scale_y_continuous(breaks = seq(0, 40, 5)) +
  scale_y_cut(breaks=c(15, 25), which=c(1,2), scales=c(1.5, 1)) +
  labs(x="Dose (mg/day)", y="Tooth Length (mm)") +
  theme_bw()

p2

Figure 2: Grouped scatter plot

3. Line plot

# Line plot
p3 <- ggplot(df, aes(x=dose, y=mean_len, color=supp)) +
  geom_line(linewidth=1) +
  geom_point(size=3) +
  scale_y_continuous(breaks = seq(0, 30, 5)) +
  scale_y_cut(breaks=c(15), which=1, scales=1.5) +
  labs(x="Dose (mg/day)", y="Mean Tooth Length (mm)") +
  theme_minimal()

p3

Figure 3: Line plot

Tip

Key Parameters:

breaks: Sets the breakpoint positions

which: Specifies the breakpoint interval (counting from the bottom up)

scales: Sets the scale factor for each interval

space: Breakpoint spacing (default 0.1)

position_dodge(): Controls the spacing between grouped bars

width is recommended to match the width parameter of geom_col.

4. More advanced charts

# More advanced charts
my_colors <- c("#1B9E77", "#D95F02")  # 来自RColorBrewer的Set2调色板
p4 <- ggplot(df, aes(x = factor(dose), y = mean_len, fill = supp)) +
  geom_col(position = position_dodge(0.8), width = 0.7, alpha = 0.9) +
  geom_errorbar(
    aes(ymin = mean_len - se_len, ymax = mean_len + se_len),
    width = 0.2, position = position_dodge(0.8)
  ) +
  geom_text(
    aes(group = supp, label = sprintf("%.1f±%.1f", mean_len, se_len)),
    position = position_dodge(0.8), vjust = -1, size = 4
  ) +
  scale_y_continuous(
    breaks = seq(0, 35, 5) # 增加上方扩展空间
  ) +
  scale_y_break(c(8,12)
  ) +
  scale_fill_manual(values = my_colors) +
  labs(x = "Dose (mg/day)", y = "Tooth Length (mm)", fill = "Supplement") +
  theme_minimal(base_size = 12) +
  theme(
    legend.position = "top",
    panel.grid.major.x = element_blank()
  )

p4

Figure 4: More advanced charts

Application

Show the practical application of visualization charts in biomedical literature. If basic charts/advanced charts are widely used in various types of biomedical literature, you can choose to display them separately.

1. Gene expression visualization (handling extreme outliers)

Figure 5: Applications of Break Plot

Changes in gene expression are shown. [1]

2. Clinical indicator distribution display (processing data of different dimensions)

Figure 6: Applications of Break Plot

The figure shows the changes in (A) [organ weight] and (B) organ-to-body weight ratio of the uteroplacenta, liver, kidney, heart, pancreas, and thymus at different stages of maternal pregnancy. [2]

Reference

[1] Xu S, et al. (2022) ggbreak: Effective Axis Break Creation in ggplot2. Journal of Open Source Software 7(74), 4301

[2] Yu D, Wang H, Shyh-Chang N. A multi-tissue metabolome atlas of primate pregnancy. Cell. 2024 Feb 1;187(3):764-781.e14. doi: 10.1016/j.cell.2023.11.043.

[3] Chen LQ, Hou BH, Mudgett MB, Frommer WB. Sugar transporters for intercellular exchange and nutrition of pathogens. Nature. 2010 Nov 25;468(7323):527-32. doi: 10.1038/nature09606.