Stream will start soon

Data analysis using R

Course overview and introduction

15.04.2021

Danny Arends
Fachgebiet Züchtungsbiologie und molekulare Tierzüchtung
Humboldt-Universität zu Berlin
 Overview

 General course announcements

 Introduction into R

 Some assignments to practice

General course announcements

 Slides will be made available online on Moodle

 Please check if you can see the course on Moodle, if not contact me

 Please attend the lecture on twitch stream

 Lectures are supported by practical exercises, spend some time on

them, if you get stuck: send me an email
 ZooM meeting every week?

 Exam date unknown at this point in time:

 Form: Online exam by written examination
General course announcements
 Questions ?

 Who followed a statistics courses ?

 Who has any programming experience ?

 What do you want to learn ?

What you will be able to do after this
course

 Format your data for R

 Load your data in R
 Do statistics on your data
 Interpret the results
 Create nice plots
 Suitable for publication
 Create your own analysis
 ‘Publish’ this as an R package for others
What you will be able to do after this
course

 Analyze your own data

 Line and dot plots
 Boxplot
 Histogram
 Heatmap
 Contour
 And many more
What you will be able to do after this
course

 Analyze your own data

 Line and dot plots
 Boxplot
 Histogram
 Heatmap
 Contour
 And many more
What you will be able to do after this
course

 Analyze your own data

 Line and dot plots
 Boxplot
 Histogram
 Heatmap
 Contour
 And many more
What you will be able to do after this
course

 Analyze your own data

 Line and dot plots
 Boxplot
 Histogram
 Heatmap
 Contour
 And many more
 Introduction into R
15.04.2021

Danny Arends
Fachgebiet Züchtungsbiologie und molekulare Tierzüchtung
Humboldt-Universität zu Berlin
 Overview

 History
 Look and Feel of R
 R as a calculator
 Types
 Indexing
 Variables
 Scripts
 Why history

 Why history in an R course ?

“We are like dwarfs on the shoulders of giants, so that we
can see more than they, and things at a greater distance . .
. we are carried high and raised up by their giant size”
- John of Salisbury

From: [Link]
 History

 2400 BC Abakus (Babylon)

 1110 BC Differential gears / south-pointing chariot (China)
 200 BC Chinese Abakus (China)
 120 BC Antikythera mechanism (Greece / Korinthe)

Abacus, public domain by its author, HB Antikythera mechanism, public domain by its author
 History

 100 Astrolabe (Unknown)

 800 Crypto, frequency analysis (Alkindus, Irak)
 1206 Castle clock (Al-Jazari, Mosul, Irak)
 1694 Leibniz wheel (Gottfried W. Leibniz, Germany)

“it is beneath the dignity of excellent men to waste their

time in calculation when any peasant could do the work
just as accurately with the aid of a machine.”
- Gottfried Leibniz
 History

• Charles Babbage
– 1791 - 1871
– Analytical Engine
• Ada Lovelace
– 1815 - 1852
– First algorithm
• Alan M. Turing
– 1912 - 1954
– Turing Machine
 History

 Build May 1941

 Z3 - The first computer
 Konrad Zuse

 Wing flutter
 Destroyed 1943

(1) Zuse Z1 replica in the German Museum of Technology in Berlin

(2) Zuse Z3-replica in het Deutsches Museum in München
(3) Konrad Zuse working in his lab (unknown)
 History

 ENIAC – Electronic Numerical Integrator And Computer

 Electronic general-purpose computer
 Turing-complete
 Digital
 Capable of being reprogrammed
 loops, branches and subroutines

 Ballistics tables
Programmers Betty Jean Jennings (left) and Fran Bilas
(right) operate ENIAC's main control panel at the Moore
School of Electrical Engineering. (U.S. Army photo from
the archives of the ARL Technical Library)
 History

 Von Neumann architecture

 1945 - John von Neumann

(1) von Neumann Architecture - licensed under the Creative Commons Attribution-Share Alike 3.0
(2) Los Alamos: Beginning of an era, 1943-1945, Los Alamos Scientific Laboratory, 1986
 Present / Future

 1976 Quantum information theory

Roman Stanisław Ingarden
 1985 Universal quantum computer
David Deutsch
 1991 Entanglement secure communication
 2011 D-Wave One (128 qubits)
 2012 Quantum Artificial Intelligence Lab D-
(512 qubits) NASA, Google, D-Wave
 2020 D-Wave Advantage™ (>5000 qubits)
 D-Wave One (128 qubits)

 Discrete optimization
 $10.000.000
 ENIAC then: $500,000
now: $6,100,000

 Lattice protein folding

D-Wave Systems, Inc. - D-Wave Systems, Inc., CC BY 3.0

 D-Wave Advantage

 Menten AI
 Protein design
 Volkswagen
 Paint shop scheduling
 Save-On-Foods
 Grocery optimization
 Amazon
 Free access for CoVID19 researchers

D-Wave Systems, Inc. - D-Wave Systems, Inc., CC BY 3.0

History - programming languages

 Low level
 Punch cards
 Assembly
 Mid level
− C - Dennis Ritchie and Ken Thompson (69 - 73)
 High Level
 Plankalkül - Konrad Zuse (43 - 45)
 AutoCode - Alick Edwards Glennie (50s)
 Algol - Committee of American and European computer scientists (Late
50s)
 LISP - John McCarthy (58)
 COBOL - Grace Hopper (59)
Too many to count, so why R?

From: [Link]
The R programming language

 Free and Open-Source

 Statistical computing
 Graphics
 Written in:
 R, Fortran, C
 Operating System agnostic
 Linear and non-linear models
 Build-in testing and help
 Many add-on packages available: 4000+
Look and Feel
 R as a calculator

 We can use R as a basic calculator:

 So typing in: 1 + 4 will yield 5
 5 / 10 will yield 0.5

 Some special operations:

 Exponents: 5 ^ 2 or 5 ** 2
 Euclidean division %/%
 Euclidean division remainder %%

 Some special numerical constants:

 Inf = infinite, NaN = Not A Number, NA = missing
Euclidean division
100 divided by 39

39 / 100 \
Euclidean division

39 / 100 \ 2 Euclidean divisor

78
---
22 Euclidean division remainder
 R as a calculator

 Build in character constants:

 LETTERS
 The 26 upper-case letters of the Roman alphabet
 letters
 The 26 lower-case letters of the Roman alphabet
 [Link]
 The three-letter abbreviations for the English month names
 [Link]
 The English names for the months of the year
 pi
 The ratio of the circumference of a circle to its diameter
 R as a calculator

 Imaginary numbers
 sqrt(-1) = NaN, but sqrt(-1+0i) = 0+1i

 Additionally basic trigonometry functions

 sin, cos, tan, asin, acos, atan

 Logarithms
 log(5) natural logarithm of 5
 log10(5) base 10 logarithm of 5
 exp(1) e^1
 Order of operations

 Operator precedence
 exponents and roots
 multiplication and division
 addition and subtraction

 PEMDAS
Please Excuse My Dear Aunt Sally

 10 - 3 x 2 = ???
 Break

 Short break
 Your session

 Everything in R is in RAM memory

 Functions to manage your session:
getwd() # Where does R save/retrieve files ?
dir() # What files are there ?
setwd() # Set another place to save/retrieve files
ls() # What is in the current environment ?

[Link]("qtl") # Install package

library(qtl) # Load a package
save(object, file = "[Link]") # Save an object as binary
[Link]("[Link]") # Save the whole environment
q("no") # quit the R session
 Help is available !

?functionname # Open the help file for function

??term # Search for a term across

?seq # Open the help for the seq function

?"+" # The + function help page

??obesity # Search anything obesity in R

 Types of data

 Logical
TRUE, FALSE

 Numeric
5, 7.9, 100.6

 Character
"one", "two", "three"

 Vector
v1 <- c(1, 2, 5.3, 6, -2,4) # Numeric vector
v2 <- c("one", "two", "three") # Character vector
v3 <- c(TRUE, TRUE, TRUE, FALSE, TRUE) # Logical vector

 Matrix
Y <- matrix(1:20, nrow = 5, ncol = 4) # 5 x 4 numeric matrix
 Working with types

 Some useful functions to deal with types

length(object) # Number of elements
str(object) # Structure of an object
class(object) # Class or type of an object
names(object) # Names

as.<typename> # Force to a certain type

is.<typename> # Is of a certain type ?
 Creating vectors & matrices

 Vector
c(object, object, ...) # Combine objects into a vector
seq(from, to, by) # A numerical sequence
rep(object, times) # Repeat an object / number

 Matrix
matrix(vector, nrow, ncol) # Create a matrix from a vector
cbind(object, object, ...) # Combine objects as columns
rbind(object, object, ...) # Combine objects as rows
 Creating vectors & matrices

 Vector
V1 <- 1:4 # 1, 2, 3, 4
V2 <- seq(1, 100, 7) # 1, 8, 15, .. 99
V3 <- rep(1, 4) # 1, 1, 1, 1
V4 <- rep("A", 4) # A, A, A, A

 Matrix
Y1 <- matrix(1:20, nrow = 5, ncol = 4) # 5 x 4 numeric matrix
Y2 <- cbind(V1,V2) # 4 x 2 numeric matrix
Y3 <- rbind(V1,V2) # 2 x 4 numeric matrix
 Vector and Matrix - Example

 Vector

Index by R

 Matrix
 Indexing a vector

 Indexing the a vector called v:

v[5]

v[2:5]

v[c(2:5,8)]
 Indexing a matrix

m[1:3, 1]
m[5, 3:6]
m[8, 7]
m[ , 9]
 Again: Types of data

 Logical
TRUE, FALSE

 Numeric
5, 7.9, 100.6

 Character
"one", "two", "three"

 Vector
a <- c(1, 2, 5.3, 6, -2,4) # Numeric vector
b <- c("one", "two", "three") # Character vector
c <- c(TRUE, TRUE, TRUE, FALSE, TRUE) # Logical vector

 Matrix
Y <- matrix(1:20, nrow = 5, ncol = 4) # 5 x 4 numeric matrix
 Advanced Types

 Data Frame (Not a matrix, can contain multiple basic types)

v1 <- c(1, 2, 3, 4)
v2 <- c("red", "white", "red", NA)
v3 <- c(TRUE, TRUE, TRUE, FALSE)
d <- [Link](v1, v2, v3)

 List (Not a vector, can contain anything)

w <- list(name="Fred", numbers = v1, age = 5.3)

 Factor
gender <- factor(c(rep("male",20), rep("female", 30)))

 Comments (use me often)

# This is a comment, R will ignore this
 Quick Type Test

 What is the type of:

 "TRUE"
 "1"
 1e+11
 0x89
 #FF00AA
 FALSE
 [Link](TRUE)
 [Link](1e+11)
 Indexing a List: [[ ]]

w <- list(name = "Fred",

numbers = c(1, 2, 3, 4),
age = 5.3,
matrix = matrix(c(1,0,0,1), nrow = 2, ncol = 2))

w[[1]][1]
w$numbers[c(2,3)]
w[[4]][,1]
w$matrix[1,]
Matrix and [Link] functions

nrow(matrix) # Number of rows

ncol(matrix) # Number of columns

rownames(matrix) # Names of the rows

colnames(matrix)

rownames(matrix) <- c("A", "B", "C") # Set row names

colnames(matrix) <- c("A", "B", "C") # Set column names

# Transpose matrix
t(matrix)
 Variables

 Variables are boxes

 You can put things in (use: <- or = )
 You can use the box without knowing what is in it
 We have already seen them:
variables <- 1.5
can <- c(TRUE, FALSE)
have <- c(1, 2, 3, 4)
many <- c(FALSE, FALSE)
names <- -5
 Code clean = create scripts

 Use a new directory for a new lecture

 Every week, use a new file
 Name the files in a logical way
 Add a header (comment section) to each file
 Name
 Date
 Purpose of the file
 Copyright
 Use a LOT of comments
 An example

# Analysis of Hardy-Weinberg equilibrium

#
# copyright (c) 2015 - HU Berlin
# written by: Danny Arends
#
# last modified Apr, 2015
# first written Feb, 2015

setwd("D:/Rcourse/assignments")
 Use a good text editor

 Windows Notepad++
 OSX Textwrangler
 Linux Use what you want

 Most important
 Code highlighting
 Support for bracket testing
 Remember

 Clean code = smart code

Questions ?

?
 Assignments

 Available from Moodle