Outline

• R data types
• R data structure
• String manipulation
• Loops
• Subsetting
• File operation
• R dates

Data types

• Basic data types in R:
  • logical
  • integer
  • double
  • character
• use typeof() function to check type of a variable.

Data types (examples)

typeof(1.1)

## [1] "double"

typeof("a")

## [1] "character"

typeof(TRUE)

## [1] "logical"

typeof(4)

## [1] "double"

typeof(4L)

## [1] "integer"

logical

• TRUE/T
• FALSE/F

if(T){
  print(TRUE)
}

## [1] TRUE

typeof(FALSE)

## [1] "logical"

is.logical(T)

## [1] TRUE

• TRUE/FALSE are recommended, because T/F can be over-written.

integer

• 1:10 represents integer sequence 1 to 10.

aseq <- 1:10 ## <- represent assign value
typeof(aseq)

## [1] "integer"

• 6L represents 6 is an integer.

aint = 6L
is.integer(aint)

## [1] TRUE

• In R, 6 generally is double instead of integer.

is.integer(6)

## [1] FALSE

Assign values

a = 1.6; print(a) ## assign 1.6 to a 

## [1] 1.6

b <- 1.6; print(b) ## assign 1.6 to b

## [1] 1.6

1.6 -> d; print(d) ## assign 1.6 to d

## [1] 1.6

Difference between = and <-

• <- can be only used for value assignment.
• = can be used for both value assignment and function argument.

double

typeof(1.4142)

## [1] "double"

is.double(pi)

## [1] TRUE

is.double(0L)

## [1] FALSE

is.double(0L + 1.5)

## [1] TRUE

character

• You can create character using single quotes or double quotes

acharacter <- "I like Biostatistical computing"
typeof(acharacter)

## [1] "character"

bcharacter <- 'You like Biostatistical computing'
is.character(bcharacter)

## [1] TRUE

• When a single quote is part of your string, you need to use double quotes.

ccharacter <- "He doesn't like Biostatistical computing"
print(ccharacter)

## [1] "He doesn't like Biostatistical computing"

1d Vector

• Vector is the basic data structure in R. Two types of vectors
  • Atomic vector: All elements of an atomic vector must be the same type
  • List: elements of a list can be of different type.
• Atomic vectors usually created with c(), short for combine

dbl_var <- c(1, 23.1, 4.2)
int_var <- c(1L, 11L, 6L)
log_var <- c(TRUE, FALSE, T, F)
chr_var <- c("I", "like Biostatistical computing")

Commonly used vector functions

Example of Vector Functions

x <- c(1,6,9,2,2,5,10)
length(x)
unique(x)
length(unique(x))
sort(x)
rev(x)
which(x==2)
which.max(x)
which.min(x)
append(x,0)
match(9,x)
y <- c(1,2,3)
union(x,y)
intersect(x,y)
setdiff(x,y)
setequal(x,y)

Statistical Vector Functions

Example of Statistical Vector Functions

avec <- c(5,2,9,3)
max(avec)

## [1] 9

which.max(avec)

## [1] 3

mean(avec)

## [1] 4.75

range(avec)

## [1] 2 9

Coercion

• All elements of an atomic vector must be the same type. Otherwise they will be coerced to the most flexible type.
• Types from least to most flexible are: logical, integer, double and character.

typeof(c("a", 1))

## [1] "character"

x <- c(FALSE, FALSE, TRUE)
as.numeric(x)

## [1] 0 0 1

as.character(x)

## [1] "FALSE" "FALSE" "TRUE"

typeof(c(1.2,1L))

## [1] "double"

How to get help

• How to get help of an R function
  • ?sum
  • help(“sum”)
• check with the best friend – google

Missing value

• Missing values are denoted by NA, which is logical vector of length 1.
• NA will always be coerced to be the correct type if used inside c()
• (optional) You can create NA of a specific type with NA_real_, NA_integer_, NA_character_

typeof(NA)

## [1] "logical"

typeof(NA_integer_)

## [1] "integer"

typeof(NA_real_)

## [1] "double"

typeof(NA_character_)

## [1] "character"

list

• Lists are different from atomic vectors because their elememts can be of any type, including lists.
• Construct lists by using list()
  
• str() function (short for structure) and gives a compact description of any R data structure.

x <- list(1:3, "a", c(TRUE, FALSE, TRUE), list(2.3, 5.9))
str(x)

## List of 4
##  $ : int [1:3] 1 2 3
##  $ : chr "a"
##  $ : logi [1:3] TRUE FALSE TRUE
##  $ :List of 2
##   ..$ : num 2.3
##   ..$ : num 5.9

x[[1]]

## [1] 1 2 3

list

• Set names for a list

x <- list(p1 = 1:3, p2 = "a", p3 = c(TRUE, FALSE, TRUE), p4 = list(2.3, 5.9))
str(x)

## List of 4
##  $ p1: int [1:3] 1 2 3
##  $ p2: chr "a"
##  $ p3: logi [1:3] TRUE FALSE TRUE
##  $ p4:List of 2
##   ..$ : num 2.3
##   ..$ : num 5.9

x[[1]]

## [1] 1 2 3

x$p1

## [1] 1 2 3

Data structure

• Can be organised by their dimensionality (1d, 2d, or nd) and whether they are homogeneous or heterogeneous.

• Note there is no 0-dimensional in R, or scalar types. Individual numbers or strings are acutally vector of length one.

Attributes

• Use to store meta-data.
  
• can be accessed individually with attr() or attributes().
• construct a new object with attributes using structure() function e.g. 
  • structure(1:10, myAttribute=“this is a vector”)

y <- 1:10
attr(y, "my_attribute") <- "This is a vector"
attr(y, "my_attribute")

## [1] "This is a vector"

attributes(y)

## $my_attribute
## [1] "This is a vector"

Attributes 2

• Three special attributes have a specific accessor function to get and set values.
  • names(): a character vector giving each element a name.
  • dim(): used to turn vectors into matrics and arrays.
  • class(): used to implement the S3 object system.

y <- c(a=1,2:10)
names(y)

##  [1] "a" ""  ""  ""  ""  ""  ""  ""  ""  ""

names(y)[2] <- 'b'
dim(y)

## NULL

dim(y) <- c(2,5)
print(y)

##      [,1] [,2] [,3] [,4] [,5]
## [1,]    1    3    5    7    9
## [2,]    2    4    6    8   10

class(y)

## [1] "matrix" "array"

factor

• A factor is a vector that only contain predefined values.
• Factors are used to store categorical data.
• Factors are built on top of character vectors using two attributes:
  • class(), “factor”, which makes them behave differently from regular character vectors.
  • levels(), which defines the set of allowed values.

x <- factor(c("a", "b", "b", 'a'))
x

## [1] a b b a
## Levels: a b

class(x)

## [1] "factor"

levels(x)

## [1] "a" "b"

factors 2

factors are very useful when there exist missing class

sex_char <- c("m", "m", "m")
sex_factor <- factor(sex_char, levels=c("m", "f"))
table(sex_char)

## sex_char
## m 
## 3

table(sex_factor)

## sex_factor
## m f 
## 3 0

Matrices

• Create a matrix with colnames and rownames

a <- matrix(1:6, ncol=3, nrow=2, dimnames = list(c("row1", "row2"),
                               c("C.1", "C.2", "C.3")))
a

##      C.1 C.2 C.3
## row1   1   3   5
## row2   2   4   6

colnames(a)

## [1] "C.1" "C.2" "C.3"

rownames(a)

## [1] "row1" "row2"

ncol(a)

## [1] 3

nrow(a)

## [1] 2

Matrices

• Adding a dim() attribute to an atomic vector

c <- 1:6
dim(c) <- c(3,2)
c

##      [,1] [,2]
## [1,]    1    4
## [2,]    2    5
## [3,]    3    6

dim(c) <- c(2,3)
c

##      [,1] [,2] [,3]
## [1,]    1    3    5
## [2,]    2    4    6

rownames(c) <- c("row1", "row2")
colnames(c) <- c("C.1", "C.2", "C.3")
c

##      C.1 C.2 C.3
## row1   1   3   5
## row2   2   4   6

Data frames

• A data frame is a very popular way of storing data in R.
• A data frame is a list of equal length vector and shares properties of both matrix and list.
  • names() and colnames() are the same thing
  • length() and ncol() are the same thing

df <- data.frame(x=1:3, y=c("a","b","c"),z=0)
str(df)

## 'data.frame':    3 obs. of  3 variables:
##  $ x: int  1 2 3
##  $ y: chr  "a" "b" "c"
##  $ z: num  0 0 0

cat(names(df), "same as", colnames(df))

## x y z same as x y z

cat(length(df), "same as", ncol(df))

## 3 same as 3

String manipulation

• some special characters:
  • ” ” for space
  • “\n” for newline
  • “\t” for tab

sentenses <- "R is a great statistical software.\n\nWe use R in the Biostatistical computing class!"
sentenses

## [1] "R is a great statistical software.\n\nWe use R in the Biostatistical computing class!"

• cat function will recognize these special characters and print to the console:

cat(sentenses)

## R is a great statistical software.
## 
## We use R in the Biostatistical computing class!

Convert to upper case or lower case

achar <- "this is a dog."
print(achar)

## [1] "this is a dog."

print(toupper(achar))

## [1] "THIS IS A DOG."

print(tolower("WWW.UFL.EDU"))

## [1] "www.ufl.edu"

Length of a string

• use nchar to count how many characters in a string instead of length

achar <- "this is a dog."
nchar(achar)

## [1] 14

length(achar)

## [1] 1

vectorizes nchar

• we can pass a vector of character to nchar

chars <- c("a dog", "a cat", "a gator")
nchar(chars)

## [1] 5 5 7

length(chars)

## [1] 3

Obtaining a substring

• take a sub-sequence of characters – use substr(), short for sub string.

chars <- "this is a dog"
substring(chars,1,1)

## [1] "t"

substring(chars,11,13)

## [1] "dog"

• replace with sub-string.

substring(chars,11,13) <- "cat"
print(chars)

## [1] "this is a cat"

strsplit

• you can split a string by a certain pattern using strsplit() function

strsplit("this is a dog", split=" ")

## [[1]]
## [1] "this" "is"   "a"    "dog"

strsplit("this is a dog", split="")

## [[1]]
##  [1] "t" "h" "i" "s" " " "i" "s" " " "a" " " "d" "o" "g"

• Note the return type is a list with only one element. strsplit can be also vectorized.

strsplit(c("this is a dog", "this is a cat", "this is a gator"), split=" ")

## [[1]]
## [1] "this" "is"   "a"    "dog" 
## 
## [[2]]
## [1] "this" "is"   "a"    "cat" 
## 
## [[3]]
## [1] "this"  "is"    "a"     "gator"

paste

• paste multiple strings

paste('this','is','a','dog', sep=" ")

## [1] "this is a dog"

paste0('this','is','a','dog')

## [1] "thisisadog"

avec <- c('this','is','a','dog')
nchar(avec)

## [1] 4 2 1 3

• collapse a vector

paste(c('this','is','a','dog'), collapse = " ")

## [1] "this is a dog"

Substituation

• use gsub to replace certain pattern within a string.

achar <- "this is a dog"
gsub(pattern = "dog",replacement="cat",x=achar) ## pattern, replacement, x

## [1] "this is a cat"

gsub("dog","cat",achar) ## pattern, replacement, x

## [1] "this is a cat"

• vectorize

chars <- c("this is a dog", "this is a cat", "this is a gator")
gsub("this","that",chars) ## pattern, replacement, x

## [1] "that is a dog"   "that is a cat"   "that is a gator"

Regular expression

• A regular expression or regex is a structured string to match specific patterns in the text.
• grep() function allows us to scan through a vector against regex

chars <- c("this is a dog", "this is a cat", "this is a gator")
grep("gator", chars)

## [1] 3

grep("this", chars)

## [1] 1 2 3

grep("gator", chars, value = T)

## [1] "this is a gator"

grep("gator", chars, invert = T)

## [1] 1 2

Regular expression 2

• match dog or cat

chars <- c("this is a dog", "this is a cat", "this is a gator")
grep("dog|cat", chars)

## [1] 1 2

Regular expression 3

• start with

chars <- c("abc", "ab", "bc")
grep("b", chars)

## [1] 1 2 3

grep("^b", chars)

## [1] 3

• end with

chars <- c("abc", "ab", "bc")
grep("ab", chars)

## [1] 1 2

grep("ab$", chars)

## [1] 2

Metacharacters

• Metacharacters are special characters with a special meaning.
• A regular expression often consists metacharacters
• Square braces are used to match anything in the braces

chars <- c("this is a dog", "this is a cat", "this is a gator")
grep("[bced]", chars)

## [1] 1 2

• dash inside square braces is used to indicate a range

chars <- c("this is a dog", "this is a cat", "this is a gator")
grep("[b-d]", chars)

## [1] 1 2

grep("[0-9]", chars)

## integer(0)

Metacharacters 2

• A caret inside braces matches anything except the followng words.
  • “[^0-9]” matches anything but a number between 0 and 9.
  • “[^aeiou]” matches anything but a lower case vowel.
• A period “.” matches to any character.

Substituation using metacharacters

• use gsub to replace certain pattern within a string.

chars <- c("this is a dog", "this is a cat", "this is a   gator")
gsub(pattern = "[aeiou]",replacement="Z",x=chars) ## pattern is vowel, replacement is Z

## [1] "thZs Zs Z dZg"     "thZs Zs Z cZt"     "thZs Zs Z   gZtZr"

gsub("[0-9]","#","a1b2") ## pattern, replacement, x

## [1] "a#b#"

Loop

• for loop

for(i in 1:10){
  cat(i," ")
}

## 1  2  3  4  5  6  7  8  9  10

• while loop

i <- 1
while(i <= 10){
  cat(i," ")
  i <- i + 1
}

## 1  2  3  4  5  6  7  8  9  10

Loop 2

• next: directly start next round of loop

for(i in 1:10){
  if(i%%2==0){
    next
  }
  cat(i," ")
}

## 1  3  5  7  9

• equivalently:

for(i in seq_len(10)){
  if(i%%2==0){
    next
  }
  cat(i," ")
}

## 1  3  5  7  9

• break: break the loop

for(i in 1:10){
  if(i%%2==0){
    break
  }
  cat(i," ")
}

## 1

Loop 3

x <- list(1:3, "a", c(TRUE, FALSE, TRUE), list(2.3, 5.9))
for(i in 1:length(x)){ 
  ax <- x[[i]]
  print(ax)
}

## [1] 1 2 3
## [1] "a"
## [1]  TRUE FALSE  TRUE
## [[1]]
## [1] 2.3
## 
## [[2]]
## [1] 5.9

for(i in seq_along(x)){ ## seq_along(x) is the same as 1:length(x) 
  ax <- x[[i]]
  print(ax)
}

## [1] 1 2 3
## [1] "a"
## [1]  TRUE FALSE  TRUE
## [[1]]
## [1] 2.3
## 
## [[2]]
## [1] 5.9

for(ax in x){ ## direct sub-element
  print(ax)
}

## [1] 1 2 3
## [1] "a"
## [1]  TRUE FALSE  TRUE
## [[1]]
## [1] 2.3
## 
## [[2]]
## [1] 5.9

Subsetting – Atomic vectors 1

example: x <- c(2.1, 4.2, 3.3, 5.4). How can we obtain a subset of this vector?

• Positive integers: return elements at the specified position.

### subseting
## atomic vectors
x <- c(2.1, 4.2, 3.3, 5.4)

# Positive integer
x[c(3,1)]

## [1] 3.3 2.1

x[order(x)] ## equivalent to sort(x)

## [1] 2.1 3.3 4.2 5.4

x[c(1,1,1)]

## [1] 2.1 2.1 2.1

Subsetting – Atomic vectors 1

example: x <- c(2.1, 4.2, 3.3, 5.4). How can we obtain a subset of this vector?

• Negative integers: omit elements at the specified positions.

# negative integer
x[-c(1, 3)]

## [1] 4.2 5.4

x[-grep(4.2, x)]

## [1] 2.1 3.3 5.4

Subsetting – Atomic vectors 3

example: x <- c(2.1, 4.2, 3.3, 5.4). How can we obtain a subset of this vector?

• Logical vectors: select elements where the corresponding logical value is TRUE.

x[c(TRUE, TRUE, FALSE, FALSE)]

## [1] 2.1 4.2

x > 3

## [1] FALSE  TRUE  TRUE  TRUE

x[x > 3]

## [1] 4.2 3.3 5.4

x[c(TRUE, TRUE, NA, FALSE)]

## [1] 2.1 4.2  NA

Subsetting – Atomic vectors 4

example: x <- c(2.1, 4.2, 3.3, 5.4). How can we obtain a subset of this vector?

• Nothing: return the original vector.

x[]

## [1] 2.1 4.2 3.3 5.4

• Zero: return a zero length vector.

x[0]

## numeric(0)

Subsetting – Atomic vectors 5

example: x <- c(2.1, 4.2, 3.3, 5.4). How can we obtain a subset of this vector?

• Character vectors: to return elements with matching names.

x <- c(a=2.1, b=4.2, c=3.3, d=5.4)

x["a"]

##   a 
## 2.1

x[letters[2:3]]

##   b   c 
## 4.2 3.3

Subsetting – Matrices and arrays

• 1d index for each dimension, separated by comma

a <- matrix(1:9, nrow=3)
colnames(a) <- c("A","B","C")
a

##      A B C
## [1,] 1 4 7
## [2,] 2 5 8
## [3,] 3 6 9

a[1:2,]

##      A B C
## [1,] 1 4 7
## [2,] 2 5 8

a[c(T,F,T), c("B","A")]

##      B A
## [1,] 4 1
## [2,] 6 3

a[,-2]

##      A C
## [1,] 1 7
## [2,] 2 8
## [3,] 3 9

Subsetting – Data frame

• Data frames possess the characteristics of both lists and matrices.

df <- data.frame(x=1:2, y=2:1, z=letters[1:2])
df[df$x==2,]

##   x y z
## 2 2 1 b

df[c("x","z")] # like a list

##   x z
## 1 1 a
## 2 2 b

df[,c("x","z")] # like a matrix

##   x z
## 1 1 a
## 2 2 b

Subsetting – simplifying vs preserving

• two types of subsetting: simplifying and preserving subsetting.
  • Simplifying subsets returns the simplest possible data structure that can represent the output.
  • Preserving subsetting keeps the structure of the output the same as the input.

Subsetting – simplifying vs preserving

x <- list(a=111, b=222)


x[[1]] ## simplified, result is numeric

## [1] 111

typeof(x[[1]])

## [1] "double"

x[1] ## preserved, result is a list

## $a
## [1] 111

typeof(x[1])

## [1] "list"

Matching

grades <- c(1,2,2,3,1)
info <- data.frame(grade=3:1, desc=c("Excellent", "Good", "Poor"), fail=c(F,F,T))
id <- match(grades, info$grade)
id

## [1] 3 2 2 1 3

info[id,]

##     grade      desc  fail
## 3       1      Poor  TRUE
## 2       2      Good FALSE
## 2.1     2      Good FALSE
## 1       3 Excellent FALSE
## 3.1     1      Poor  TRUE

File I/O

Read in txt/csv files

• read in locally

setwd("~/Desktop") ## set your working directory
burnData <- read.csv("burn.csv", row.names = 1)

• read directly from URL

burnData <- read.csv("https://caleb-huo.github.io/teaching/data/Burn/burn.csv", row.names = 1)

• read only the first two lines

burnData <- read.csv("burn.csv", row.names = 1, nrows = 2)

• skip the first three lines

burnData <- read.csv("burn.csv", row.names = 1, skip = 3)

Delimiter

• comma:

read.csv()

• tab:

read.table()

• specify your own:

read.delim(, delim=";")

Also pay attention to the arguments such as header, row.names

Save txt/csv files

write.csv(burnData, file = "myBurnData.csv")
write.table(burnData, file = "myBurnData.txt")
write.table(burnData, file = "myBurnData.txt", append = TRUE)

Read in line by line

fileNameFull <- 'https://caleb-huo.github.io/teaching/data/Burn/burn.csv'
con  <- file(fileNameFull, open = "r")

while (length(oneLine <- readLines(con, n = 1, warn = FALSE)) > 0) {
    aline = strsplit(oneLine, ",")[[1]]
    print(aline)
} 
close(con) ## remember to close files

Save R objects part 1 (save)

If you take a long time to obtain your result. How to save your result so in the future, you won’t bother re-run them again?

• Create a list for these variables
• Save the list by save()

a <- 1:4
b <- 2:5
ans <- a * b
result <- list(a=a, b=b, ans=ans)
save(result,file="myResult.rdata")

• load the list by load()

load("myResult.rdata")
result2 <- get(load("myResult.rdata"))

Save R objects part 1 (saveRDS)

If you take a long time to obtain your result. How to save your result so in the future, you won’t bother re-run them again?

• Create a list for these variables
• Save the list by save()

a <- 1:4
b <- 2:5
ans <- a * b
result <- list(a=a, b=b, ans=ans)
saveRDS(result,file="myResult.rdata")

• load the list by load()

result2 <- readRDS("myResult.rdata")

Date

Dates are represented as the number of days since 1970-01-01, with negative values for earlier dates.

mydates <- as.Date(c("2017-09-11", "2012-12-17", "1970-01-01"))
# number of days in between 
days <- mydates[1] - mydates[2]
days

## Time difference of 1729 days

as.numeric(mydates)

## [1] 17420 15691     0

Sys.Date( ) ## returns today's date.

## [1] "2023-09-04"

date() ## returns the current date and time.

## [1] "Mon Sep  4 17:48:19 2023"

Date format

today <- Sys.Date()
format(today, format="%B %d %Y")

## [1] "September 04 2023"

Date Conversion

• as.Date() function: to convert character data to dates

# convert date info in format 'mm/dd/yyyy'
strDates <- c("01/05/1995", "08/16/1995")
dates <- as.Date(strDates, "%m/%d/%Y")

• The default format is yyyy-mm-dd

as.Date(c("2007-06-22", "2004-02-13"))

## [1] "2007-06-22" "2004-02-13"

• Date to Character

as.character(dates)

## [1] "1995-01-05" "1995-08-16"

• Date to Numeric (1970-01-01 as 0)

as.numeric(dates)

## [1] 9135 9358

References

• http://adv-r.had.co.nz
• http://www.stat.cmu.edu/~ryantibs/statcomp/
• https://www.statmethods.net/input/dates.html