如何用ggplot2绘制带基因标签的Fold Change柱状图？

Hey there! Let's get that bar chart working properly for your gene expression data. I'll break down what went wrong and walk you through the correct code step by step.

What Was Off With Your Original Code #

Your initial attempt had two critical issues that prevented it from working as expected:

• You used a quoted string 'Fold Change' for the x aesthetic—ggplot interprets this as a single static category instead of referencing the actual variable in your dataset.
• geom_bar() defaults to counting observations (making a frequency chart), but you want to plot the exact numerical values of Fold Change. To do this, you need to explicitly set stat = "identity".

Step 1: Clean Your Data (Handle NA Values) #

First, let's deal with the NA entry for CARD18. ggplot will skip these by default, but it's cleaner to filter them out explicitly:

# Load ggplot2 if you haven't already
library(ggplot2)

# Remove rows where Fold Change is NA
clean_profile <- Patient11GeneExpressionProfile[!is.na(Patient11GeneExpressionProfile$`Fold Change`), ]

Step 2: Build the Correct Bar Chart #

Now we'll map Gene to the x-axis (so each bar gets a unique gene label) and Fold Change to the y-axis (to set the height of each bar). We'll also add a small formatting tweak to keep x-axis labels readable:

ggplot(clean_profile, aes(x = Gene, y = `Fold Change`)) +
  geom_bar(stat = "identity", fill = "steelblue") + # Use actual Fold Change values for bar height
  labs(
    title = "Fold Change per Gene (Patient 11)",
    x = "Gene",
    y = "Fold Change"
  ) +
  theme(axis.text.x = element_text(angle = 45, hjust = 1)) # Rotate x labels to avoid overlap

Optional Tweaks (For Better Clarity) #

If you want to highlight genes that are upregulated (Fold Change > 1) or downregulated, you can add color coding and a reference line:

ggplot(clean_profile, aes(x = Gene, y = `Fold Change`, fill = `Fold Change` > 1)) +
  geom_bar(stat = "identity") +
  geom_hline(yintercept = 1, linetype = "dashed", color = "darkred") + # Reference line for no change
  labs(
    title = "Fold Change per Gene (Patient 11)",
    x = "Gene",
    y = "Fold Change",
    fill = "Upregulated"
  ) +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

内容的提问来源于stack exchange，提问作者Marcus