09_R_introduction

By Yan Li

PhD in Bioinformatics, University of Liverpool

Approaching R

R: an open source programming language for statistical analysis

  • Get familar with the interface of R-studio
  • Use R-studio manage package
  • Use R to perform a t-test
  • Use R to visualise data

R-studio

R-studio

Set working directory

  • Click Session - Set Working Directory - Choose Directory
  • Or simply use cmd setwd("your/working/directory")

Import data and check

# Read data
voles <- read.csv("Voles.txt", sep='\t')

# Check the data
names(voles)
head(voles)
summary(voles)

Subset

volesmh <- subset(voles, Sex=="Male" & IL1b_253=="G:A")
summary (volesmh)

Simple plot

barplot()
boxplot()
hist()
plot()
pie()

...

Simple statistic

mean()
median()
sd()
sum()
t.test()

Change CRAN mirror

  • Click Tools - Global options - Packages
  • Change CRAN repo to TUNA

cran

Package

install.packages("ggplot2")
require("ggplot2")

Workshop

  • Copy the dataset worms.txt to you laptop.
  • Test this hypothesis: Whether there is a statistically significant difference of worm density between in damp field and in dry field.

Reading