Loading......

Password

Plotting

Section 1. R Graphics Fundamentals

Why Learn R Graphics?

Visualize data patterns
Create publication-quality figures
Master fundamental plotting techniques

Basic Workflow (3 Steps)

1. Create Canvas

png("my_plot.png")
# or
pdf("my_plot.pdf")

2. Draw Graphics

plot(1,1)
plot(c(1,2),c(2,3))
boxplot(rnorm(10),rnorm(10))

3. Save Output

# Save 

dev.off()

Graphic Devices

Screen Preview

plot(rnorm(100))  # Automatically displays in RStudio

File Saving

pdf("output.pdf")  # Start PDF device
hist(mtcars$mpg)   # Draw plot
dev.off()          # Save file

Color Parameter (`col`)

Basic Usage

plot(1:5, col=c("red","green","blue","cyan","magenta"), pch=16, cex=3)

Named colors: “red”, “blue”, etc
RGB: col="#FF0000" for pure red

Point Shapes (`pch`)

Common Shapes

# Plot different point styles
plot(1:6, rep(1,6), pch=16:21, col=rainbow(6), cex=3)

Numbers 0-25 = different shapes
16= filled circle, 17=filled triangle

Line Types (`lty`)

Basic Patterns

# Create empty plot
plot(1, type="n", xlim=c(0,6), ylim=c(0,7))

# Draw different lines
for(i in 1:6) {
  abline(h=i, lty=i-1, lwd=2)
}

Parameter Combination

Customized Plot

plot(mtcars$wt, mtcars$mpg,
     col=ifelse(mtcars$am==1, "red", "blue"),
     pch=ifelse(mtcars$cyl==4, 16, 17),
     lty=2)

Question: how to draw this plot using “if … else …” ?

Key Takeaways

Always follow: Create → Draw → Save
Manage devices properly
Master basic parameters: col/pch/lty

Section 2. Basic Plotting with plot()

Today’s Roadmap

Basic usage of plot()
Controlling graph types with type parameter
Live coding examples

What is `plot()`?

Simplest function to create graphs
Works with vectors/data frames
Automatically chooses axes ranges

# Minimal example
plot(1:5)

Scatterplot Basics (Numeric Vectors)

Create arithmetic sequence with : operator:

x <- 1:10
y <- 1:10
plot(x, y)

Formula Style Plotting

Use y ~ x syntax with datasets:

# Using built-in mtcars data
plot(mpg ~ wt, data = mtcars)

Understanding the `type` Parameter

Controls how points are displayed:

Code	Meaning
“p”	Points
“l”	Lines
“b”	Both points and lines

Type “p” - Point Plot (Default)

x <- 1:10
y <- rnorm(10)  # Random numbers
plot(x, y, type = "p")

Type “l” - Line Plot

Connect points with lines:

x <- seq(0, 2*pi, length=50)
y <- sin(x)
plot(x, y, type = "l")

Type “b” - Both Points and Lines

Best of both worlds:

x <- c(1,3,5,7,9)
y <- c(2,4,1,8,5)
plot(x, y, type = "b")

Let’s Make a Fancy Sine Wave

Combine parameters for better visualization:

x <- seq(0, 2*pi, length=50)
y <- sin(x)
plot(x, y, 
     type = "b", 
     col = "blue",
     main = "Sine Wave Demo",
     xlab = "Angle (radians)",
     ylab = "sin(x)")

Pro Tip: Plot Customization

You can add these to any plot:

main = "Title"    # Add title
col = "red"       # Change color
pch = 17          # Change point shape
cex = 2           # Make points bigger

Live Demo Time!

Let’s create together:

Simple scatterplot of 20 random numbers
Line plot of cosine function
Combined plot with custom colors

Section 3. Layering Graphic Elements

Why Layer Graphics?

Build complex plots step-by-step
Combine multiple elements in one visualization
Customize existing plots with additional information

Basic Plot Setup

# Create base plot
plot(mtcars$mpg, mtcars$hp, 
     main = "Car Performance", 
     xlab = "MPG", ylab = "Horsepower",
     pch = 16, col = "gray")

1. Adding Points with `points()`

# Highlight high-performance cars
special_cars <- mtcars[mtcars$hp > 300, ]
points(special_cars$mpg, special_cars$hp, 
       pch = 17,  # Triangle shape
       col = "red",
       cex = 1.5) # Size

What’s happening:

pch=17: Triangle plotting symbol
col="red": Makes points stand out
cex=1.5: 50% larger than default

2. Adding Lines with `lines()`

# Add linear regression line
model <- lm(hp ~ mpg, data = mtcars) # linear regression will be taught next week
x_values <- seq(10, 35, length.out = 100)
y_values <- predict(model, newdata = data.frame(mpg = x_values)) 
lines(x_values, y_values, 
      col = "blue", 
      lwd = 2,    # Line width
      lty = 2)    # Dashed line

Note: Always calculate prediction values first!

3. Drawing Rectangles with `rect()`

# Highlight unusual MPG range
rect(xleft = 15, ybottom = 50, 
     xright = 20, ytop = 350,
     border = "red", 
     col = rgb(1, 0, 0, 0.2)) # Transparent red

Parameters:

xleft, xright: X-axis boundaries
ybottom, ytop: Y-axis boundaries
rgb(): Creates transparent color

4. Adding Segments with `segments()`

# Add mean reference lines to boxplot
boxplot(mpg ~ cyl, data = mtcars, col = "lightblue")
cyl_means <- aggregate(mpg ~ cyl, mtcars, mean) # aggregate function is use to stat X based on Y
segments(x0 = 1:3 - 0.4, y0 = cyl_means$mpg,
         x1 = 1:3 + 0.4, y1 = cyl_means$mpg,
         col = "darkred", lwd = 3)

Breakdown:

x0/x1: Horizontal line positions
y0/y1: Vertical positions (same for horizontal line)
lwd=3: Thick line for visibility

Layering Order Matters!

Always create base plot first
Add elements in desired z-order:
- Background elements first
- Main data points next
- Annotations last
Use par(new=TRUE) carefully (not recommended for beginners)

Try This!

Combine all elements we learned:

plot(mtcars$mpg, mtcars$hp)
points(special_cars$mpg, special_cars$hp, col="red")
lines(x_values, y_values, col="blue")
rect(15, 50, 20, 350, border="red")
segments(10, 150, 30, 150, col="green")

Experiment with colors and positions!

Section 04. Essential Statistical Plots

Let’s Make Heatmaps!

What is a Heatmap?

A picture that shows patterns in numbers using colors
Example: Gene expression levels (high=red, low=white)

Basic Heatmap Code

# Create sample data
gene_data <- matrix(rnorm(100, mean=10), nrow=10)
colnames(gene_data) <- paste("Patient", 1:10)
rownames(gene_data) <- paste("Gene", LETTERS[1:10])

# Simple heatmap
heatmap(gene_data)

Understanding the Output

Colors show values (legend on left)
Tree diagrams show grouping patterns
Rows=Genes, Columns=Patients

Customizing Heatmaps

Remove the tree diagrams:

heatmap(gene_data, 
        Rowv = NA,  # No row clustering
        Colv = NA)  # No column clustering

Boxplots Made Easy

What’s a Boxplot?

Shows data distribution through 5 numbers:
📦 Minimum, Q1, Median, Q3, Maximum
Perfect for comparing groups!

Boxplot with Iris Flowers

# Use built-in iris data
boxplot(Petal.Length ~ Species, 
        data = iris,
        col = c("pink", "lightblue", "yellowgreen"))

Understanding the Parts

Middle line = Median
Box edges = Q1 and Q3
Whiskers = Data range (Q1－1.5IQR & Q3+1.5IQR; IQR: Interquartile Range)
Points = Outliers

Making It Pretty

Add colors and labels:

boxplot(Petal.Length ~ Species,
        data = iris,
        main = "Petal Length Comparison",
        xlab = "Iris Species",
        ylab = "Length (cm)",
        col = heat.colors(3))

Why These Matter?

Heatmaps help see:
🧬 Gene expression patterns
🦠 Microbial communities

Boxplots help compare:
🌷 Flower measurements
🏥 Medical test results

Try These Modifications!

Experiment with colors in heatmaps:

heatmap(gene_data, col = terrain.colors(50))

Change boxplot orientation:

boxplot(Petal.Length ~ Species, data=iris, horizontal=TRUE)

Remember These Keys

Heatmap essentials:
🔥 heatmap() needs a numeric matrix
🎨 Use col= to change colors

Boxplot basics:
📦 Formula syntax: y ~ group
🌈 Colors make comparisons easier

Section 5. Practical Visualization Cases

Today’s Roadmap

Biomedical visualization cases
Graphical output techniques
Hands-on plotting examples

Why Visualization Matters

In biomedical research:

# Good visualization can reveal:
- Dose-response relationships 📈
- Group differences 🧪🔬
- Data quality issues ⚠️

Case 1: Dose-Response Curve

Scenario: Test drug efficacy at different concentrations

# Sample data
doses <- c(0, 10, 25, 50, 100)
response <- c(5, 15, 40, 65, 85)

# Basic plot
plot(doses, response, 
     type = "b", 
     main = "Drug Dose-Response Curve",
     xlab = "Dose (mg/mL)", 
     ylab = "Effect (%)",
     col = "darkred",
     pch = 19)

Beautifying Our Plot

Add professional touches:

# Add gridlines
grid(col = "lightgray")

# Add confidence intervals
arrows(doses, response-5, doses, response+5, 
       length=0.05, angle=90, code=3)

Case 2: Metabolic Comparison

Task: Compare cholesterol levels between groups

# Generate sample data
set.seed(123)
control <- rnorm(50, mean = 180, sd = 20)
treatment <- rnorm(50, mean = 160, sd = 15)

# Create boxplot
boxplot(list(Control=control, Treatment=treatment),
        col = c("#FFC107", "#2196F3"),
        main = "Cholesterol Levels Comparison",
        ylab = "mg/dL")

Boxplot Interpretation Guide

# Key elements:
- Central line: Median 
- Box: 25%-75% percentile 
- Whiskers: 1.5×IQR 
- Outliers: Beyond whiskers ⚠

Saving as PNG

Best for web/quick sharing

png("my_plot.png", width=800, height=600)

# Plotting commands
plot(doses, response)
dev.off() # IMPORTANT!

🔑 Pro Tip: Always close device with dev.off()!

PDF Output

Best for publications/editing

pdf("figure.pdf", 
    width=8, # inches
    height=6)
    
# Plotting commands
boxplot(control, treatment)
dev.off()

📄 Advantage: Vector graphics = Infinite zoom!

Common Mistakes to Avoid

Forgetting dev.off() -> Empty files ❌
Wrong plot dimensions -> Cropped labels 📏
Saving while plot window is open -> Conflicts 💥
Overwriting files -> Use unique names 🏷️

Let’s Practice Together!

Create a combined plot:

# Set up PDF output
pdf("combined_plots.pdf", width=10, height=5)

# Split screen
par(mfrow=c(1,2))

plot(doses, response, main="Dose Response")
boxplot(list(C=control, T=treatment), main="Cholesterol")

dev.off()

Key Takeaways

Basic plots reveal biological patterns
Always customize plots for clarity
Choose PNG/PDF based on needs
Device management is crucial
Combine plots for comparison

Loading......

Password

Plotting

Section 1. R Graphics Fundamentals

Why Learn R Graphics?

Basic Workflow (3 Steps)

1. Create Canvas

2. Draw Graphics

3. Save Output

Graphic Devices

Screen Preview

File Saving

Color Parameter (col)

Basic Usage

Point Shapes (pch)

Common Shapes

Line Types (lty)

Basic Patterns

Parameter Combination

Customized Plot

Question: how to draw this plot using “if … else …” ?

Key Takeaways

Section 2. Basic Plotting with plot()

Today’s Roadmap

What is plot()?

Scatterplot Basics (Numeric Vectors)

Formula Style Plotting

Understanding the type Parameter

Type “p” - Point Plot (Default)

Type “l” - Line Plot

Type “b” - Both Points and Lines

Let’s Make a Fancy Sine Wave

Pro Tip: Plot Customization

Live Demo Time!

Section 3. Layering Graphic Elements

Why Layer Graphics?

Basic Plot Setup

1. Adding Points with points()

2. Adding Lines with lines()

3. Drawing Rectangles with rect()

4. Adding Segments with segments()

Layering Order Matters!

Try This!

Section 04. Essential Statistical Plots

Let’s Make Heatmaps!

What is a Heatmap?

Basic Heatmap Code

Understanding the Output

Customizing Heatmaps

Boxplots Made Easy

What’s a Boxplot?

Boxplot with Iris Flowers

Understanding the Parts

Making It Pretty

Why These Matter?

Try These Modifications!

Remember These Keys

Today’s Roadmap

Why Visualization Matters

Case 1: Dose-Response Curve

Beautifying Our Plot

Case 2: Metabolic Comparison

Boxplot Interpretation Guide

Saving as PNG

PDF Output

Common Mistakes to Avoid

Let’s Practice Together!

Key Takeaways

Color Parameter (`col`)

Point Shapes (`pch`)

Line Types (`lty`)

What is `plot()`?

Understanding the `type` Parameter

1. Adding Points with `points()`

2. Adding Lines with `lines()`

3. Drawing Rectangles with `rect()`

4. Adding Segments with `segments()`