Now that we know about data types and structures, let’s see how to manipulate them.
# Basic arithmetic
2 + 5[1] 710 - 3[1] 74 * 8[1] 32100 / 4[1] 25# Order of operations (like standard math)
5 + 2 * 3 # Multiplication first[1] 11(5 + 2) * 3 # Parentheses first[1] 21# Built-in mathematical functions
sqrt(16) # Square root[1] 4log(10) # Natural logarithm[1] 2.302585log10(100) # Base-10 logarithm[1] 2Used to ask TRUE/FALSE questions about our data. Essential for filtering.
> : Greater than< : Less than>=: Greater than or equal to<=: Less than or equal to==: Exactly equal to (TWO equal signs! Very common mistake to use just one =)!=: Not equal to& : AND (both sides must be TRUE)| : OR (at least one side must be TRUE)! : NOT (reverses TRUE to FALSE, FALSE to TRUE)yield <- 5.2
min_acceptable_yield <- 5.0
variety <- "ICARDA_Gold"
# Comparisons
yield > min_acceptable_yield # Is yield acceptable? TRUE[1] TRUEvariety == "Local_Check" # Is it the local check? FALSE[1] FALSEvariety != "Local_Check" # Is it NOT the local check? TRUE[1] TRUE# On vectors
plot_yields <- c(5.2, 4.5, 6.1, 5.5)
plot_yields > 5.0 # Which plots yielded above 5.0? [TRUE FALSE TRUE TRUE][1] TRUE FALSE TRUE TRUEplot_varieties <- c("ICARDA_Gold", "Local_Check", "ICARDA_RustResist",
"ICARDA_Gold")
# Which plots are ICARDA_Gold? [TRUE FALSE FALSE TRUE]
plot_varieties == "ICARDA_Gold"[1] TRUE FALSE FALSE TRUE# Combining conditions
# Find plots where yield > 5.0 AND variety is ICARDA_Gold
(plot_yields > 5.0) & (plot_varieties == "ICARDA_Gold") # [TRUE FALSE FALSE TRUE][1] TRUE FALSE FALSE TRUE# Find plots where yield > 6.0 OR variety is Local_Check
(plot_yields > 6.0) | (plot_varieties == "Local_Check") # [FALSE TRUE TRUE FALSE][1] FALSE TRUE TRUE FALSEMany R operations are vectorized, meaning they automatically apply to each element of a vector without needing you to write a loop. This makes R code concise and efficient. We’ve already seen this with arithmetic (plot_yields + 0.5) and comparisons (plot_yields > 5.0).
Functions like mean(), sum(), min(), max(), sd() (standard deviation), length() also work naturally on vectors:
plot_yields <- c(5.2, 4.5, 6.1, 5.5)
mean(plot_yields)[1] 5.325sd(plot_yields)[1] 0.6652067sum(plot_yields > 5.0) # How many plots yielded > 5.0? (TRUE=1, FALSE=0)[1] 3length(plot_yields) # How many plots?[1] 4This is crucial for selecting specific data from your tables.
Let’s use the trial_data data frame from the previous section:
trial_data <- data.frame(
PlotID = c("A101", "A102", "B101", "B102"),
Variety = factor(c("ICARDA_Gold", "Local_Check", "ICARDA_RustResist",
"ICARDA_Gold")),
Yield_kg_plot = c(5.2, 4.5, 6.1, 5.5),
Is_Resistant = c(TRUE, FALSE, TRUE, TRUE)
)Accessing Columns: Use $ (most common) or [[ ]]. trial_data$Variety or trial_data[["Variety"]]
Accessing Rows/Columns/Cells using [row, column]:
# Get the value in Row 2, Column 3
trial_data[2, 3] # Should be 4.5[1] 4.5# Get the entire Row 1 (returns a data frame)
trial_data[1, ] PlotID Variety Yield_kg_plot Is_Resistant
1 A101 ICARDA_Gold 5.2 TRUE# Get the entire Column 2 (Variety column, returns a vector/factor)
trial_data[, 2][1] ICARDA_Gold Local_Check ICARDA_RustResist ICARDA_Gold
Levels: ICARDA_Gold ICARDA_RustResist Local_Check# Get Columns 1 and 3 (PlotID and Yield)
trial_data[, c(1, 3)] # Use c() for multiple column indices PlotID Yield_kg_plot
1 A101 5.2
2 A102 4.5
3 B101 6.1
4 B102 5.5trial_data[, c("PlotID", "Yield_kg_plot")] # Can also use column names PlotID Yield_kg_plot
1 A101 5.2
2 A102 4.5
3 B101 6.1
4 B102 5.5row part of the square brackets.# Select rows where Yield_kg_plot is greater than 5.0
high_yield_plots <- trial_data[trial_data$Yield_kg_plot > 5.0, ]
print(high_yield_plots) PlotID Variety Yield_kg_plot Is_Resistant
1 A101 ICARDA_Gold 5.2 TRUE
3 B101 ICARDA_RustResist 6.1 TRUE
4 B102 ICARDA_Gold 5.5 TRUE# Select rows where Variety is "ICARDA_Gold"
icarda_gold_plots <- trial_data[trial_data$Variety == "ICARDA_Gold", ]
print(icarda_gold_plots) PlotID Variety Yield_kg_plot Is_Resistant
1 A101 ICARDA_Gold 5.2 TRUE
4 B102 ICARDA_Gold 5.5 TRUE# Select rows where Variety is "ICARDA_Gold" AND yield > 5.0
# (We generated the logical vector for this earlier)
condition <- (trial_data$Variety == "ICARDA_Gold") &
(trial_data$Yield_kg_plot > 5.0)
print(condition) # Shows [TRUE FALSE FALSE TRUE][1] TRUE FALSE FALSE TRUEselected_plots <- trial_data[condition, ]
print(selected_plots) PlotID Variety Yield_kg_plot Is_Resistant
1 A101 ICARDA_Gold 5.2 TRUE
4 B102 ICARDA_Gold 5.5 TRUE# Select rows where the variety is resistant
resistant_plots <- trial_data[trial_data$Is_Resistant == TRUE, ]
# Or just trial_data[trial_data$Is_Resistant, ]
print(resistant_plots) PlotID Variety Yield_kg_plot Is_Resistant
1 A101 ICARDA_Gold 5.2 TRUE
3 B101 ICARDA_RustResist 6.1 TRUE
4 B102 ICARDA_Gold 5.5 TRUEThe apply() family of functions lets us apply a function to the rows or columns in a matrix or data frame, a list or a vector.
# Imagine we have a data matrix of plot yield values for different varieties.
# Each row represents a variety and each column a yield measurement for each trial
plot_trials <- matrix(c(5.2, 4.5, 6.1, 5.5, 4, 6.6, 7, 5.1, 5.3),
nrow = 3, ncol = 3)
# We calculate the mean for each variety (each row)
# 1 means function is run on rows, 2 would mean function is run on columns
apply(plot_trials, 1, mean)[1] 5.900000 4.533333 6.000000Exercise: Select the data for the ‘Local_Check’ variety from the trial_data data frame.
trial_data[trial_data$Variety == "Local_Check",] PlotID Variety Yield_kg_plot Is_Resistant
2 A102 Local_Check 4.5 FALSE