Tutorial 1: R Fundamentals & Your First Biomedical Dataset

Introduction to R for Biomedical Science

Introduction

Learning Objectives:

By the end of this tutorial, you will be able to:

• Navigate the RStudio interface and create an R project
• Write, annotate, and run basic R code
• Create and manipulate vectors and data frames using biomedical data
• Import a CSV dataset and perform exploratory analysis
• Filter, select, and transform data using dplyr
• Export a cleaned dataset to CSV

---

1. Setting Up Your RStudio Project (10 minutes)

Why Projects Matter

Projects automatically set your working directory and keep all files organised. This prevents the classic “cannot open file” error.

Step-by-step instructions:

1. Open RStudio
2. Go to File → New Project → New Directory → New Project
3. Name the project r_biomedical_intro
4. Choose a sensible location (e.g. your Desktop or Documents folder)
5. Click Create Project — RStudio will restart
6. Go to File → New File → R Script — save it as tutorial_01.R

Generate the Dataset

Before we begin, let’s create our clinical dataset. Run this code once:

# Generate clinical dataset for Tutorial 1
set.seed(42)
n <- 80

clinical_data <- data.frame(
  patient_id  = paste0("PT", sprintf("%03d", 1:n)),
  age         = round(rnorm(n, mean = 52, sd = 14)),
  sex         = sample(c("Male", "Female"), n, replace = TRUE),
  bmi         = round(rnorm(n, mean = 26.5, sd = 4.2), 1),
  treatment   = sample(c("Control", "Drug_A", "Drug_B"), n, replace = TRUE),
  wbc_count   = round(rnorm(n, mean = 7.2, sd = 1.8), 2),   # white blood cells x10^9/L
  systolic_bp = round(rnorm(n, mean = 130, sd = 18)),
  response    = round(rnorm(n, mean = 50, sd = 20), 1)      # % symptom reduction
)

# Save to CSV
write.csv(clinical_data, "clinical_data.csv", row.names = FALSE)

# Display first few rows
head(clinical_data)

##   patient_id age    sex  bmi treatment wbc_count systolic_bp response
## 1      PT001  71   Male 20.2   Control      8.79         147     16.5
## 2      PT002  44 Female 20.3    Drug_A      9.44         143     85.7
## 3      PT003  57 Female 27.0   Control      5.31         127      4.5
## 4      PT004  61   Male 22.3    Drug_B      6.00         120     18.2
## 5      PT005  58 Female 26.5    Drug_A      8.52         144     45.1
## 6      PT006  51 Female 24.7   Control      6.28         125     42.9

---

2. Writing Basic R Code (10 minutes)

Comments and Documentation

Lines starting with # are comments — R ignores them. Always comment your code!

# -------------------------------------------------------
# Tutorial 1: R Fundamentals
# Author: Your Name
# Date: Today's Date
# Description: Exploring a clinical dataset in R
# -------------------------------------------------------

# This is a comment — it documents what the code does
# Your future self will thank you for writing clear comments

Arithmetic Operations

R can be used as a powerful calculator:

# Basic arithmetic
2 + 3          # addition

## [1] 5

10 - 4         # subtraction

## [1] 6

6 * 7          # multiplication

## [1] 42

100 / 4        # division

## [1] 25

2^8            # exponentiation (2 to the power of 8)

## [1] 256

sqrt(144)      # square root — R has many built-in functions

## [1] 12

Variable Assignment

Use <- to store values in variables (objects):

# Variable assignment
# Use <- to store a value in a variable
# Think of <- as "gets": "x gets the value 42"
x <- 42
x              # typing the variable name prints its value

## [1] 42

patient_count <- 80
mean_age <- 52.4
study_name <- "Cardio Trial 2024"   # text (strings) go in quotes

# Print multiple variables
print(paste("Study:", study_name, "- Mean age:", mean_age))

## [1] "Study: Cardio Trial 2024 - Mean age: 52.4"

Note: R is case-sensitive. Age and age are different objects.

---

3. Vectors (10 minutes)

A vector is the fundamental unit of R — a collection of values of the same type.

Creating Vectors

Use c() (combine/concatenate) to create vectors:

# A vector of white blood cell counts (x10^9/L) for 6 patients
wbc <- c(5.2, 7.8, 11.3, 4.9, 8.1, 6.7)

# A vector of patient ages
ages <- c(34, 56, 23, 67, 45, 71)

# A vector of treatment groups (character vector)
treatment <- c("Control", "Drug_A", "Drug_A", "Control", "Drug_B", "Drug_B")

# Display
wbc

## [1]  5.2  7.8 11.3  4.9  8.1  6.7

Exploring Vectors

# Basic statistics
length(wbc)       # how many elements?

## [1] 6

sum(wbc)          # total

## [1] 44

mean(wbc)         # average

## [1] 7.333333

median(wbc)       # median

## [1] 7.25

sd(wbc)           # standard deviation

## [1] 2.341509

range(wbc)        # min and max

## [1]  4.9 11.3

Indexing and Subsetting

R counts from 1 (not 0 like Python):

# Accessing elements by position
wbc[1]            # first element

## [1] 5.2

wbc[3]            # third element

## [1] 11.3

wbc[c(1, 3, 5)]   # first, third, and fifth elements

## [1]  5.2 11.3  8.1

wbc[2:4]          # elements 2 through 4

## [1]  7.8 11.3  4.9

# Logical indexing: find elements that meet a condition
wbc > 8           # returns TRUE/FALSE for each element

## [1] FALSE FALSE  TRUE FALSE  TRUE FALSE

wbc[wbc > 8]      # returns only values greater than 8

## [1] 11.3  8.1

# Which patients have elevated WBC?
# Normal adult WBC range is roughly 4.5 to 11.0 x10^9/L
high_wbc <- wbc[wbc > 11.0]
high_wbc

## [1] 11.3

Vectorised Operations

Apply operations to all elements at once — no loops needed:

# Vectorised operations (a superpower of R)
wbc * 1000        # convert to cells per microlitre

## [1]  5200  7800 11300  4900  8100  6700

wbc - mean(wbc)   # centre the data (subtract mean from each value)

## [1] -2.1333333  0.4666667  3.9666667 -2.4333333  0.7666667 -0.6333333

---

4. Data Frames & Importing Data (15 minutes)

A data frame is like a spreadsheet table — rows are observations, columns are variables.

Loading Packages

We’ll use the tidyverse collection of packages:

# Load the tidyverse (includes ggplot2, dplyr, readr, tidyr, and more)
library(tidyverse)

Importing CSV Data

# Load the clinical dataset we created earlier
clinical_data <- read_csv("clinical_data.csv")

Exploring Data Frames

# First look at your data
head(clinical_data)         # first 6 rows

## # A tibble: 6 × 8
##   patient_id   age sex      bmi treatment wbc_count systolic_bp response
##   <chr>      <dbl> <chr>  <dbl> <chr>         <dbl>       <dbl>    <dbl>
## 1 PT001         71 Male    20.2 Control        8.79         147     16.5
## 2 PT002         44 Female  20.3 Drug_A         9.44         143     85.7
## 3 PT003         57 Female  27   Control        5.31         127      4.5
## 4 PT004         61 Male    22.3 Drug_B         6            120     18.2
## 5 PT005         58 Female  26.5 Drug_A         8.52         144     45.1
## 6 PT006         51 Female  24.7 Control        6.28         125     42.9

tail(clinical_data)         # last 6 rows

## # A tibble: 6 × 8
##   patient_id   age sex      bmi treatment wbc_count systolic_bp response
##   <chr>      <dbl> <chr>  <dbl> <chr>         <dbl>       <dbl>    <dbl>
## 1 PT075         44 Male    22.5 Drug_A        10.2          155     47.3
## 2 PT076         60 Male    31.1 Drug_A         8.13         135     53.9
## 3 PT077         63 Male    28.2 Drug_B         7.11         135     31.9
## 4 PT078         58 Female  29   Drug_B         4.9          105     43.7
## 5 PT079         40 Female  34.1 Control        6.32         132     27  
## 6 PT080         37 Female  27   Drug_A         9.52         137     12.5

glimpse(clinical_data)      # structure: column names, types, sample values

## Rows: 80
## Columns: 8
## $ patient_id  <chr> "PT001", "PT002", "PT003", "PT004", "PT005", "PT006", "PT0…
## $ age         <dbl> 71, 44, 57, 61, 58, 51, 73, 51, 80, 51, 70, 84, 33, 48, 50…
## $ sex         <chr> "Male", "Female", "Female", "Male", "Female", "Female", "M…
## $ bmi         <dbl> 20.2, 20.3, 27.0, 22.3, 26.5, 24.7, 23.9, 18.0, 21.4, 27.3…
## $ treatment   <chr> "Control", "Drug_A", "Control", "Drug_B", "Drug_A", "Contr…
## $ wbc_count   <dbl> 8.79, 9.44, 5.31, 6.00, 8.52, 6.28, 6.43, 6.08, 7.56, 9.90…
## $ systolic_bp <dbl> 147, 143, 127, 120, 144, 125, 110, 119, 124, 151, 161, 147…
## $ response    <dbl> 16.5, 85.7, 4.5, 18.2, 45.1, 42.9, 55.4, 59.1, -3.8, 67.3,…

summary(clinical_data)      # statistical summary of each column

##   patient_id             age            sex                 bmi       
##  Length:80          Min.   :10.00   Length:80          Min.   :18.00  
##  Class :character   1st Qu.:43.00   Class :character   1st Qu.:23.70  
##  Mode  :character   Median :54.00   Mode  :character   Median :26.20  
##                     Mean   :52.31                      Mean   :26.02  
##                     3rd Qu.:61.25                      3rd Qu.:28.57  
##                     Max.   :84.00                      Max.   :34.10  
##   treatment           wbc_count       systolic_bp       response    
##  Length:80          Min.   : 1.170   Min.   : 87.0   Min.   :-3.80  
##  Class :character   1st Qu.: 5.938   1st Qu.:118.8   1st Qu.:31.88  
##  Mode  :character   Median : 7.090   Median :132.0   Median :47.40  
##                     Mean   : 7.137   Mean   :130.3   Mean   :48.38  
##                     3rd Qu.: 8.220   3rd Qu.:141.5   3rd Qu.:64.88  
##                     Max.   :11.970   Max.   :161.0   Max.   :91.90

# Dimensions
dim(clinical_data)          # rows × columns

## [1] 80  8

nrow(clinical_data)         # number of rows

## [1] 80

ncol(clinical_data)         # number of columns

## [1] 8

colnames(clinical_data)     # column names

## [1] "patient_id"  "age"         "sex"         "bmi"         "treatment"  
## [6] "wbc_count"   "systolic_bp" "response"

Accessing Columns

# Using $ to pull out a column as a vector
clinical_data$age

##  [1] 71 44 57 61 58 51 73 51 80 51 70 84 33 48 50 61 48 15 18 70 48 27 50 69 79
## [26] 46 48 27 58 43 58 62 66 43 59 28 41 40 18 53 55 47 63 42 33 58 41 72 46 61
## [51] 57 41 74 61 53 56 62 53 10 56 47 55 60 72 42 70 57 67 65 62 37 51 61 39 44
## [76] 60 63 58 40 37

mean(clinical_data$age)

## [1] 52.3125

# Count observations per group
table(clinical_data$treatment)

## 
## Control  Drug_A  Drug_B 
##      24      33      23

table(clinical_data$sex)

## 
## Female   Male 
##     49     31

---

5. Data Wrangling with dplyr (10 minutes)

The Pipe Operator: |>

The pipe operator means “take this thing, THEN do this to it”:

# Without pipe (nested, hard to read):
mean(sqrt(c(4, 9, 16, 25)))

## [1] 3.5

# With pipe (left to right, readable):
c(4, 9, 16, 25) |> sqrt() |> mean()

## [1] 3.5

# Read it aloud as "and then"

Note: %>% (from magrittr) works the same as |> (base R). Use whichever you prefer.

filter(): Keep Rows

Keep only rows that meet certain conditions:

# Keep only female patients
female_patients <- clinical_data |>
  filter(sex == "Female")

# Keep patients over 60 in the Drug_A group
older_drug_a <- clinical_data |>
  filter(age > 60, treatment == "Drug_A")

# Keep patients with elevated WBC (above normal range)
elevated_wbc <- clinical_data |>
  filter(wbc_count > 11.0)

# Display result
head(elevated_wbc)

## # A tibble: 2 × 8
##   patient_id   age sex     bmi treatment wbc_count systolic_bp response
##   <chr>      <dbl> <chr> <dbl> <chr>         <dbl>       <dbl>    <dbl>
## 1 PT043         63 Male   27.2 Drug_B         11.6         144     36.2
## 2 PT046         58 Male   32.5 Drug_A         12.0         114     76.9

select(): Keep Columns

Select specific columns you need:

# Select just the columns needed for a sub-analysis
patient_summary <- clinical_data |>
  select(patient_id, age, sex, treatment, response)

head(patient_summary)

## # A tibble: 6 × 5
##   patient_id   age sex    treatment response
##   <chr>      <dbl> <chr>  <chr>        <dbl>
## 1 PT001         71 Male   Control       16.5
## 2 PT002         44 Female Drug_A        85.7
## 3 PT003         57 Female Control        4.5
## 4 PT004         61 Male   Drug_B        18.2
## 5 PT005         58 Female Drug_A        45.1
## 6 PT006         51 Female Control       42.9

mutate(): Create or Transform Columns

Create new columns or modify existing ones:

# Create a new column: WBC converted to cells per microlitre
clinical_data <- clinical_data |>
  mutate(wbc_per_ul = wbc_count * 1000)

# Create a new column: categorise BMI
clinical_data <- clinical_data |>
  mutate(bmi_category = case_when(
    bmi < 18.5              ~ "Underweight",
    bmi >= 18.5 & bmi < 25 ~ "Normal",
    bmi >= 25 & bmi < 30   ~ "Overweight",
    bmi >= 30               ~ "Obese"
  ))

# Check the result
table(clinical_data$bmi_category)

## 
##      Normal       Obese  Overweight Underweight 
##          30          13          36           1

Chaining Multiple Steps

The pipe operator shines when combining operations:

# Chain multiple operations together
drug_response_summary <- clinical_data |>
  filter(treatment != "Control") |>                    # exclude control group
  select(patient_id, treatment, response, age) |>
  filter(age >= 18 & age <= 65) |>                     # working-age adults
  mutate(high_responder = response > 60)               # flag high responders

head(drug_response_summary)

## # A tibble: 6 × 5
##   patient_id treatment response   age high_responder
##   <chr>      <chr>        <dbl> <dbl> <lgl>         
## 1 PT002      Drug_A        85.7    44 TRUE          
## 2 PT004      Drug_B        18.2    61 FALSE         
## 3 PT005      Drug_A        45.1    58 FALSE         
## 4 PT008      Drug_A        59.1    51 FALSE         
## 5 PT013      Drug_A        42.4    33 FALSE         
## 6 PT015      Drug_B        64.4    50 TRUE

Exporting Data

# Save the cleaned dataset
write_csv(clinical_data, "clinical_data_cleaned.csv")

---

6. Formative Exercise

Instructions: The dataset blood_markers.csv (generated below) contains haematology data from 40 patients. Complete the following tasks:

# Generate the exercise dataset
set.seed(99)
n2 <- 40
blood_markers <- data.frame(
  id          = paste0("P", 1:n2),
  haemoglobin = round(rnorm(n2, mean = 13.5, sd = 1.8), 1),  # g/dL
  platelets   = round(rnorm(n2, mean = 250, sd = 60)),       # x10^9/L
  condition   = sample(c("Anaemia", "Healthy", "Polycythaemia"), n2, replace = TRUE),
  hospital    = sample(c("City", "Royal", "St. Mary's"), n2, replace = TRUE)
)
write.csv(blood_markers, "blood_markers.csv", row.names = FALSE)

Tasks:

1. Import blood_markers.csv and display its structure using at least two exploration functions
2. How many patients are in each condition group? (Hint: table())
3. Filter to keep only patients with haemoglobin below 12.0 g/dL (clinically anaemic threshold)
4. Create a new column called platelet_status that labels:
   • platelets below 150 as "Low"
   • 150–400 as "Normal"
   • above 400 as "High"
5. Select only id, haemoglobin, condition, and platelet_status, and export to blood_markers_cleaned.csv

---

7. Discussion Questions

1. Reproducibility: When would you use filter() vs. manually removing rows from a CSV in Excel? What are the advantages of the scripted approach?

2. Data cleaning: You have a dataset where the sex column contains entries like "M", "male", "MALE", and "Male". How would you standardise this in R? Why does this matter?

3. Random seeds: Why is it important to use set.seed() when generating or sampling data? What could go wrong if two researchers ran the same simulation without it?

---

Further Reading & Resources

• R for Data Science (free online): https://r4ds.hadley.nz — Chapters 1–5 cover everything in this tutorial
• RStudio Cheatsheet — Base R: https://rstudio.github.io/cheatsheets/base-r.pdf
• dplyr Cheatsheet: https://rstudio.github.io/cheatsheets/data-transformation.pdf
• UCI Heart Disease Dataset (practice): https://archive.ics.uci.edu/dataset/45/heart+disease

---

Session Info

sessionInfo()

## R version 4.5.1 (2025-06-13 ucrt)
## Platform: x86_64-w64-mingw32/x64
## Running under: Windows 11 x64 (build 22631)
## 
## Matrix products: default
##   LAPACK version 3.12.1
## 
## locale:
## [1] LC_COLLATE=English_United Kingdom.utf8 
## [2] LC_CTYPE=English_United Kingdom.utf8   
## [3] LC_MONETARY=English_United Kingdom.utf8
## [4] LC_NUMERIC=C                           
## [5] LC_TIME=English_United Kingdom.utf8    
## 
## time zone: Europe/London
## tzcode source: internal
## 
## attached base packages:
## [1] stats     graphics  grDevices utils     datasets  methods   base     
## 
## other attached packages:
##  [1] lubridate_1.9.4 forcats_1.0.1   stringr_1.6.0   dplyr_1.1.4    
##  [5] purrr_1.2.0     readr_2.1.6     tidyr_1.3.1     tibble_3.3.0   
##  [9] ggplot2_4.0.1   tidyverse_2.0.0
## 
## loaded via a namespace (and not attached):
##  [1] bit_4.6.0          gtable_0.3.6       jsonlite_2.0.0     crayon_1.5.3      
##  [5] compiler_4.5.1     tidyselect_1.2.1   parallel_4.5.1     jquerylib_0.1.4   
##  [9] scales_1.4.0       yaml_2.3.12        fastmap_1.2.0      R6_2.6.1          
## [13] generics_0.1.4     knitr_1.50         bslib_0.9.0        pillar_1.11.1     
## [17] RColorBrewer_1.1-3 tzdb_0.5.0         rlang_1.1.6        utf8_1.2.6        
## [21] stringi_1.8.7      cachem_1.1.0       xfun_0.55          sass_0.4.10       
## [25] S7_0.2.1           bit64_4.6.0-1      timechange_0.3.0   cli_3.6.5         
## [29] withr_3.0.2        magrittr_2.0.4     digest_0.6.39      grid_4.5.1        
## [33] vroom_1.6.7        rstudioapi_0.17.1  hms_1.1.4          lifecycle_1.0.4   
## [37] vctrs_0.6.5        evaluate_1.0.5     glue_1.8.0         farver_2.1.2      
## [41] rmarkdown_2.30     tools_4.5.1        pkgconfig_2.0.3    htmltools_0.5.9