dblogr/

R Tutorial

An introduction to R

Introduction

This tutorial is will introduce the reader to , a free, open-source statistical computing environment often used with RStudio, a integrated development environment for .

R Project Logo
R Project Logo

Calculator

can be used as a super awesome calculator

# 5 + 3 = 8
5 + 3 
## [1] 8
# 24 / (1 + 2) = 8
24 / (1 + 2) 
## [1] 8
# 2 * 2 * 2 = 8
2^3 
## [1] 8
# 8 * 8 = 64
sqrt(64) 
## [1] 8
# -log10(0.05 / 5000000) = 8
-log10(0.05 / 5000000) 
## [1] 8

Functions

has many useful built in functions

1:10
##  [1]  1  2  3  4  5  6  7  8  9 10
as.character(1:10)
##  [1] "1"  "2"  "3"  "4"  "5"  "6"  "7"  "8"  "9"  "10"
rep(1:2, times = 5)
##  [1] 1 2 1 2 1 2 1 2 1 2
rep(1:5, times = 2)
##  [1] 1 2 3 4 5 1 2 3 4 5
rep(1:5, each = 2)
##  [1] 1 1 2 2 3 3 4 4 5 5
rep(1:5, length.out = 7)
## [1] 1 2 3 4 5 1 2
seq(5, 50, by = 5)
##  [1]  5 10 15 20 25 30 35 40 45 50
seq(5, 50, length.out = 5)
## [1]  5.00 16.25 27.50 38.75 50.00
paste(1:10, 20:30, sep = "-")
##  [1] "1-20"  "2-21"  "3-22"  "4-23"  "5-24"  "6-25"  "7-26"  "8-27"  "9-28"  "10-29" "1-30"
paste(1:10, collapse = "-")
## [1] "1-2-3-4-5-6-7-8-9-10"
paste0("x", 1:10)
##  [1] "x1"  "x2"  "x3"  "x4"  "x5"  "x6"  "x7"  "x8"  "x9"  "x10"
min(1:10)
## [1] 1
max(1:10)
## [1] 10
range(1:10)
## [1]  1 10
mean(1:10)
## [1] 5.5
sd(1:10)
## [1] 3.02765

Custom Functions

Users can also create their own functions

customFunction1 <- function(x, y) {
  z <- 100 * x / (x + y)
  paste(z, "%")
}
customFunction1(x = 10, y = 90)
## [1] "10 %"
customFunction2 <- function(x) {
  mymin <- mean(x - sd(x))
  mymax <- mean(x) + sd(x)
  print(paste("Min =", mymin))
  print(paste("Max =", mymax))
}
customFunction2(x = 1:10)
## [1] "Min = 2.47234964590251"
## [1] "Max = 8.52765035409749"

for loops and if else statements

xx <- NULL #creates and empty object
for(i in 1:10) {
  xx[i] <- i*3
}
xx
##  [1]  3  6  9 12 15 18 21 24 27 30
xx %% 2 #gives the remainder when divided by 2
##  [1] 1 0 1 0 1 0 1 0 1 0
for(i in 1:length(xx)) {
  if((xx[i] %% 2) == 0) {
    print(paste(xx[i],"is Even"))
  } else { 
      print(paste(xx[i],"is Odd")) 
    }
}
## [1] "3 is Odd"
## [1] "6 is Even"
## [1] "9 is Odd"
## [1] "12 is Even"
## [1] "15 is Odd"
## [1] "18 is Even"
## [1] "21 is Odd"
## [1] "24 is Even"
## [1] "27 is Odd"
## [1] "30 is Even"
# or
ifelse(xx %% 2 == 0, "Even", "Odd")
##  [1] "Odd"  "Even" "Odd"  "Even" "Odd"  "Even" "Odd"  "Even" "Odd"  "Even"
paste(xx, ifelse(xx %% 2 == 0, "is Even", "is Odd"))
##  [1] "3 is Odd"   "6 is Even"  "9 is Odd"   "12 is Even" "15 is Odd"  "18 is Even" "21 is Odd"  "24 is Even" "27 is Odd"  "30 is Even"

Objects

Information can be stored in user defined objects, in multiple forms:

  • c(): a string of values
  • matrix(): a two dimensional matrix in one format
  • data.frame(): a two dimensional matrix where each column can be a different format
  • list():

A string…

xc <- 1:10
xc
##  [1]  1  2  3  4  5  6  7  8  9 10
xc <- c(1,2,3,4,5,6,7,8,9,10)
xc
##  [1]  1  2  3  4  5  6  7  8  9 10

A matrix…

xm <- matrix(1:100, nrow = 10, ncol = 10, byrow = T)
xm
##       [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10]
##  [1,]    1    2    3    4    5    6    7    8    9    10
##  [2,]   11   12   13   14   15   16   17   18   19    20
##  [3,]   21   22   23   24   25   26   27   28   29    30
##  [4,]   31   32   33   34   35   36   37   38   39    40
##  [5,]   41   42   43   44   45   46   47   48   49    50
##  [6,]   51   52   53   54   55   56   57   58   59    60
##  [7,]   61   62   63   64   65   66   67   68   69    70
##  [8,]   71   72   73   74   75   76   77   78   79    80
##  [9,]   81   82   83   84   85   86   87   88   89    90
## [10,]   91   92   93   94   95   96   97   98   99   100
xm <- matrix(1:100, nrow = 10, ncol = 10, byrow = F)
xm
##       [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10]
##  [1,]    1   11   21   31   41   51   61   71   81    91
##  [2,]    2   12   22   32   42   52   62   72   82    92
##  [3,]    3   13   23   33   43   53   63   73   83    93
##  [4,]    4   14   24   34   44   54   64   74   84    94
##  [5,]    5   15   25   35   45   55   65   75   85    95
##  [6,]    6   16   26   36   46   56   66   76   86    96
##  [7,]    7   17   27   37   47   57   67   77   87    97
##  [8,]    8   18   28   38   48   58   68   78   88    98
##  [9,]    9   19   29   39   49   59   69   79   89    99
## [10,]   10   20   30   40   50   60   70   80   90   100

A data frame…

xd <- data.frame(
  x1 = c("aa","bb","cc","dd","ee",
         "ff","gg","hh","ii","jj"),
  x2 = 1:10,
  x3 = c(1,1,1,1,1,2,2,2,3,3),
  x4 = rep(c(1,2), times = 5),
  x5 = rep(1:5, times = 2),
  x6 = rep(1:5, each = 2),
  x7 = seq(5, 50, by = 5),
  x8 = log10(1:10),
  x9 = (1:10)^3,
  x10 = c(T,T,T,F,F,T,T,F,F,F)
)
xd
##    x1 x2 x3 x4 x5 x6 x7        x8   x9   x10
## 1  aa  1  1  1  1  1  5 0.0000000    1  TRUE
## 2  bb  2  1  2  2  1 10 0.3010300    8  TRUE
## 3  cc  3  1  1  3  2 15 0.4771213   27  TRUE
## 4  dd  4  1  2  4  2 20 0.6020600   64 FALSE
## 5  ee  5  1  1  5  3 25 0.6989700  125 FALSE
## 6  ff  6  2  2  1  3 30 0.7781513  216  TRUE
## 7  gg  7  2  1  2  4 35 0.8450980  343  TRUE
## 8  hh  8  2  2  3  4 40 0.9030900  512 FALSE
## 9  ii  9  3  1  4  5 45 0.9542425  729 FALSE
## 10 jj 10  3  2  5  5 50 1.0000000 1000 FALSE

A list…

xl <- list(xc, xm, xd)
xl[[1]]
##  [1]  1  2  3  4  5  6  7  8  9 10
xl[[2]]
##       [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10]
##  [1,]    1   11   21   31   41   51   61   71   81    91
##  [2,]    2   12   22   32   42   52   62   72   82    92
##  [3,]    3   13   23   33   43   53   63   73   83    93
##  [4,]    4   14   24   34   44   54   64   74   84    94
##  [5,]    5   15   25   35   45   55   65   75   85    95
##  [6,]    6   16   26   36   46   56   66   76   86    96
##  [7,]    7   17   27   37   47   57   67   77   87    97
##  [8,]    8   18   28   38   48   58   68   78   88    98
##  [9,]    9   19   29   39   49   59   69   79   89    99
## [10,]   10   20   30   40   50   60   70   80   90   100
xl[[3]]
##    x1 x2 x3 x4 x5 x6 x7        x8   x9   x10
## 1  aa  1  1  1  1  1  5 0.0000000    1  TRUE
## 2  bb  2  1  2  2  1 10 0.3010300    8  TRUE
## 3  cc  3  1  1  3  2 15 0.4771213   27  TRUE
## 4  dd  4  1  2  4  2 20 0.6020600   64 FALSE
## 5  ee  5  1  1  5  3 25 0.6989700  125 FALSE
## 6  ff  6  2  2  1  3 30 0.7781513  216  TRUE
## 7  gg  7  2  1  2  4 35 0.8450980  343  TRUE
## 8  hh  8  2  2  3  4 40 0.9030900  512 FALSE
## 9  ii  9  3  1  4  5 45 0.9542425  729 FALSE
## 10 jj 10  3  2  5  5 50 1.0000000 1000 FALSE

Selecting Data

xc[5] # 5th element in xc
## [1] 5
xd$x3[5] # 5th element in col "x3"
## [1] 1
xd[5,"x3"] # row 5, col "x3"
## [1] 1
xd$x3 # all of col "x3"
##  [1] 1 1 1 1 1 2 2 2 3 3
xd[,"x3"] # all rows, col "x3"
##  [1] 1 1 1 1 1 2 2 2 3 3
xd[3,] # row 3, all cols
##   x1 x2 x3 x4 x5 x6 x7        x8 x9  x10
## 3 cc  3  1  1  3  2 15 0.4771213 27 TRUE
xd[c(2,4),c("x4","x5")] # rows 2 & 4, cols "x4" & "x5"
##   x4 x5
## 2  2  2
## 4  2  4
xl[[3]]$x1 # 3rd object in the list, col "x1
##  [1] "aa" "bb" "cc" "dd" "ee" "ff" "gg" "hh" "ii" "jj"

regexpr

xx <- data.frame(Name = c("Item 1 (detail 1)",
                          "Item 20 (detail 20)",
                          "Item 300 (detail 300)"),
                 Item = NA,
                 Detail = NA)
xx$Detail <- substr(xx$Name, regexpr("\\(", xx$Name)+1, regexpr("\\)", xx$Name)-1)
xx$Item <- substr(xx$Name, 1, regexpr("\\(", xx$Name)-2)
xx
##                    Name     Item     Detail
## 1     Item 1 (detail 1)   Item 1   detail 1
## 2   Item 20 (detail 20)  Item 20  detail 20
## 3 Item 300 (detail 300) Item 300 detail 300

Data Formats

Data can also be saved in many formats:

  • numeric
  • integer
  • character
  • factor
  • logical
xd$x3 <- as.character(xd$x3)
xd$x3
##  [1] "1" "1" "1" "1" "1" "2" "2" "2" "3" "3"
xd$x3 <- as.numeric(xd$x3)
xd$x3
##  [1] 1 1 1 1 1 2 2 2 3 3
xd$x3 <- as.factor(xd$x3)
xd$x3
##  [1] 1 1 1 1 1 2 2 2 3 3
## Levels: 1 2 3
xd$x3 <- factor(xd$x3, levels = c("3","2","1"))
xd$x3
##  [1] 1 1 1 1 1 2 2 2 3 3
## Levels: 3 2 1
xd$x10
##  [1]  TRUE  TRUE  TRUE FALSE FALSE  TRUE  TRUE FALSE FALSE FALSE
as.numeric(xd$x10) # TRUE = 1, FALSE = 0
##  [1] 1 1 1 0 0 1 1 0 0 0
sum(xd$x10)
## [1] 5

Internal structure of an object can be checked with str()

str(xc) # c()
##  num [1:10] 1 2 3 4 5 6 7 8 9 10
str(xm) # matrix()
##  int [1:10, 1:10] 1 2 3 4 5 6 7 8 9 10 ...
str(xd) # data.frame()
## 'data.frame':    10 obs. of  10 variables:
##  $ x1 : chr  "aa" "bb" "cc" "dd" ...
##  $ x2 : int  1 2 3 4 5 6 7 8 9 10
##  $ x3 : Factor w/ 3 levels "3","2","1": 3 3 3 3 3 2 2 2 1 1
##  $ x4 : num  1 2 1 2 1 2 1 2 1 2
##  $ x5 : int  1 2 3 4 5 1 2 3 4 5
##  $ x6 : int  1 1 2 2 3 3 4 4 5 5
##  $ x7 : num  5 10 15 20 25 30 35 40 45 50
##  $ x8 : num  0 0.301 0.477 0.602 0.699 ...
##  $ x9 : num  1 8 27 64 125 216 343 512 729 1000
##  $ x10: logi  TRUE TRUE TRUE FALSE FALSE TRUE ...
str(xl) # list()
## List of 3
##  $ : num [1:10] 1 2 3 4 5 6 7 8 9 10
##  $ : int [1:10, 1:10] 1 2 3 4 5 6 7 8 9 10 ...
##  $ :'data.frame':    10 obs. of  10 variables:
##   ..$ x1 : chr [1:10] "aa" "bb" "cc" "dd" ...
##   ..$ x2 : int [1:10] 1 2 3 4 5 6 7 8 9 10
##   ..$ x3 : num [1:10] 1 1 1 1 1 2 2 2 3 3
##   ..$ x4 : num [1:10] 1 2 1 2 1 2 1 2 1 2
##   ..$ x5 : int [1:10] 1 2 3 4 5 1 2 3 4 5
##   ..$ x6 : int [1:10] 1 1 2 2 3 3 4 4 5 5
##   ..$ x7 : num [1:10] 5 10 15 20 25 30 35 40 45 50
##   ..$ x8 : num [1:10] 0 0.301 0.477 0.602 0.699 ...
##   ..$ x9 : num [1:10] 1 8 27 64 125 216 343 512 729 1000
##   ..$ x10: logi [1:10] TRUE TRUE TRUE FALSE FALSE TRUE ...

Packages

Additional libraries can be installed and loaded for use.

install.packages("scales")
library(scales)
xx <- data.frame(Values = 1:10)
xx$Rescaled <- rescale(x = xx$Values, to = c(1,30))
xx
##    Values  Rescaled
## 1       1  1.000000
## 2       2  4.222222
## 3       3  7.444444
## 4       4 10.666667
## 5       5 13.888889
## 6       6 17.111111
## 7       7 20.333333
## 8       8 23.555556
## 9       9 26.777778
## 10     10 30.000000

libraries can also be used without having to load them

scales::rescale(1:10, to = c(1,30))
##  [1]  1.000000  4.222222  7.444444 10.666667 13.888889 17.111111 20.333333 23.555556 26.777778 30.000000

Data Wrangling

R for Data Science - https://r4ds.had.co.nz/

xx <- data.frame(Group = c("X","X","Y","Y","Y","X","X","X","Y","Y"),
                 Data1 = 1:10, 
                 Data2 = seq(10, 100, by = 10))
xx$NewData1 <- xx$Data1 + xx$Data2
xx$NewData2 <- xx$Data1 * 1000
xx
##    Group Data1 Data2 NewData1 NewData2
## 1      X     1    10       11     1000
## 2      X     2    20       22     2000
## 3      Y     3    30       33     3000
## 4      Y     4    40       44     4000
## 5      Y     5    50       55     5000
## 6      X     6    60       66     6000
## 7      X     7    70       77     7000
## 8      X     8    80       88     8000
## 9      Y     9    90       99     9000
## 10     Y    10   100      110    10000
xx$Data1 < 5 # which are less than 5
##  [1]  TRUE  TRUE  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE
xx[xx$Data1 < 5,]
##   Group Data1 Data2 NewData1 NewData2
## 1     X     1    10       11     1000
## 2     X     2    20       22     2000
## 3     Y     3    30       33     3000
## 4     Y     4    40       44     4000
xx[xx$Group == "X", c("Group","Data2","NewData1")]
##   Group Data2 NewData1
## 1     X    10       11
## 2     X    20       22
## 6     X    60       66
## 7     X    70       77
## 8     X    80       88

Data wrangling with tidyverse and pipes (%>%)

library(tidyverse) # install.packages("tidyverse")
xx <- data.frame(Group = c("X","X","Y","Y","Y","Y","Y","X","X","X")) %>%
  mutate(Data1 = 1:10, 
         Data2 = seq(10, 100, by = 10),
         NewData1 = Data1 + Data2,
         NewData2 = Data1 * 1000)
xx
##    Group Data1 Data2 NewData1 NewData2
## 1      X     1    10       11     1000
## 2      X     2    20       22     2000
## 3      Y     3    30       33     3000
## 4      Y     4    40       44     4000
## 5      Y     5    50       55     5000
## 6      Y     6    60       66     6000
## 7      Y     7    70       77     7000
## 8      X     8    80       88     8000
## 9      X     9    90       99     9000
## 10     X    10   100      110    10000
filter(xx, Data1 < 5)
##   Group Data1 Data2 NewData1 NewData2
## 1     X     1    10       11     1000
## 2     X     2    20       22     2000
## 3     Y     3    30       33     3000
## 4     Y     4    40       44     4000
xx %>% filter(Data1 < 5)
##   Group Data1 Data2 NewData1 NewData2
## 1     X     1    10       11     1000
## 2     X     2    20       22     2000
## 3     Y     3    30       33     3000
## 4     Y     4    40       44     4000
xx %>% filter(Group == "X") %>% 
  select(Group, NewColName=Data2, NewData1)
##   Group NewColName NewData1
## 1     X         10       11
## 2     X         20       22
## 3     X         80       88
## 4     X         90       99
## 5     X        100      110
xs <- xx %>% 
  group_by(Group) %>% 
  summarise(Data2_mean = mean(Data2),
            Data2_sd = sd(Data2),
            NewData2_mean = mean(NewData2),
            NewData2_sd = sd(NewData2))
xs
## # A tibble: 2 × 5
##   Group Data2_mean Data2_sd NewData2_mean NewData2_sd
##   <chr>      <dbl>    <dbl>         <dbl>       <dbl>
## 1 X             60     41.8          6000       4183.
## 2 Y             50     15.8          5000       1581.
xx %>% left_join(xs, by = "Group")
##    Group Data1 Data2 NewData1 NewData2 Data2_mean Data2_sd NewData2_mean NewData2_sd
## 1      X     1    10       11     1000         60 41.83300          6000    4183.300
## 2      X     2    20       22     2000         60 41.83300          6000    4183.300
## 3      Y     3    30       33     3000         50 15.81139          5000    1581.139
## 4      Y     4    40       44     4000         50 15.81139          5000    1581.139
## 5      Y     5    50       55     5000         50 15.81139          5000    1581.139
## 6      Y     6    60       66     6000         50 15.81139          5000    1581.139
## 7      Y     7    70       77     7000         50 15.81139          5000    1581.139
## 8      X     8    80       88     8000         60 41.83300          6000    4183.300
## 9      X     9    90       99     9000         60 41.83300          6000    4183.300
## 10     X    10   100      110    10000         60 41.83300          6000    4183.300

Read/Write data

xx <- read.csv("data_r_tutorial.csv")
write.csv(xx, "data_r_tutorial.csv", row.names = F)

For excel sheets, the package readxl can be used to read in sheets of data.

library(readxl) # install.packages("readxl")
xx <- read_xlsx("data_r_tutorial.xlsx", sheet = "Data")

Tidy Data

yy <- xx %>%
  group_by(Name, Location) %>%
  summarise(Mean_DTF = round(mean(DTF),1)) %>% 
  arrange(Location)
yy
## # A tibble: 9 × 3
## # Groups:   Name [3]
##   Name          Location            Mean_DTF
##   <chr>         <chr>                  <dbl>
## 1 CDC Maxim AGL Jessore, Bangladesh     86.7
## 2 ILL 618 AGL   Jessore, Bangladesh     79.3
## 3 Laird AGL     Jessore, Bangladesh     76.8
## 4 CDC Maxim AGL Metaponto, Italy       134. 
## 5 ILL 618 AGL   Metaponto, Italy       138. 
## 6 Laird AGL     Metaponto, Italy       137. 
## 7 CDC Maxim AGL Saskatoon, Canada       52.5
## 8 ILL 618 AGL   Saskatoon, Canada       47  
## 9 Laird AGL     Saskatoon, Canada       56.8
yy <- yy %>% spread(key = Location, value = Mean_DTF)
yy
## # A tibble: 3 × 4
## # Groups:   Name [3]
##   Name          `Jessore, Bangladesh` `Metaponto, Italy` `Saskatoon, Canada`
##   <chr>                         <dbl>              <dbl>               <dbl>
## 1 CDC Maxim AGL                  86.7               134.                52.5
## 2 ILL 618 AGL                    79.3               138.                47  
## 3 Laird AGL                      76.8               137.                56.8
yy <- yy %>% gather(key = TraitName, value = Value, 2:4)
yy
## # A tibble: 9 × 3
## # Groups:   Name [3]
##   Name          TraitName           Value
##   <chr>         <chr>               <dbl>
## 1 CDC Maxim AGL Jessore, Bangladesh  86.7
## 2 ILL 618 AGL   Jessore, Bangladesh  79.3
## 3 Laird AGL     Jessore, Bangladesh  76.8
## 4 CDC Maxim AGL Metaponto, Italy    134. 
## 5 ILL 618 AGL   Metaponto, Italy    138. 
## 6 Laird AGL     Metaponto, Italy    137. 
## 7 CDC Maxim AGL Saskatoon, Canada    52.5
## 8 ILL 618 AGL   Saskatoon, Canada    47  
## 9 Laird AGL     Saskatoon, Canada    56.8
yy <- yy %>% spread(key = Name, value = Value)
yy
## # A tibble: 3 × 4
##   TraitName           `CDC Maxim AGL` `ILL 618 AGL` `Laird AGL`
##   <chr>                         <dbl>         <dbl>       <dbl>
## 1 Jessore, Bangladesh            86.7          79.3        76.8
## 2 Metaponto, Italy              134.          138.        137. 
## 3 Saskatoon, Canada              52.5          47          56.8

Base Plotting

We will start with some basic plotting using the base function plot()

# A basic scatter plot
plot(x = xd$x8, y = xd$x9)

# Adjust color and shape of the points
plot(x = xd$x8, y = xd$x9, col = "darkred", pch = 0)

plot(x = xd$x8, y = xd$x9, col = xd$x4, pch = xd$x4)

# Adjust plot type 
plot(x = xd$x8, y = xd$x9, type = "line")

# Adjust linetype
plot(x = xd$x8, y = xd$x9, type = "line", lty = 2)

# Plot lines and points
plot(x = xd$x8, y = xd$x9, type = "both")

Now lets create some random and normally distributed data to make some more complicated plots

# 100 random uniformly distributed numbers ranging from 0 - 100
ru <- runif(100, min = 0, max = 100)
ru
##   [1] 82.128105 25.671108 20.857380  6.987244 44.535229 57.439018 66.387953 76.771348 10.457316 50.624719 30.176817 15.572481  9.629041 31.565060 85.309647
##  [16] 42.826766 67.174919  1.536771 20.366357 20.778852 98.025970 80.113802  5.903874 55.539444 98.764508 89.996384 74.310003 21.675698  5.822534 80.367322
##  [31] 10.925103 86.677461 98.318522 60.201699 61.515534 12.860652 73.760295  4.827495 80.669155  8.645009 44.330766 23.171595 39.901974 31.236912 76.898981
##  [46] 31.661671 23.726180  2.674868 31.859265 37.690767 74.111909 80.620398 80.833172 44.351958 69.391265 79.600119 23.848310  3.123521 15.427447  9.320959
##  [61] 83.582061 21.123495 65.183362  2.049110 90.153031 42.336720 91.736905 77.226256 66.053051 50.628892 44.421337  4.114649 33.960568 76.080229 36.900464
##  [76] 74.752274 72.994119 32.578509 41.499422 67.894886 72.182660 75.989280 54.702159 92.080373 95.683817 41.845799 85.469616 44.819617 27.779676 36.653669
##  [91] 82.832142 41.776966 30.927743 82.130361 88.123850 36.092642 37.125076 62.295121 68.873884 26.782792
plot(x = ru)

order(ru)
##   [1]  18  64  48  58  72  38  29  23   4  40  60  13   9  31  36  59  12  19  20   3  62  28  42  47  57   2 100  89  11  93  44  14  46  49  78  73  96  90  75
##  [40]  97  50  43  79  92  86  66  16  41  54  71   5  88  10  70  83  24   6  34  35  98  63  69   7  17  80  99  55  81  77  37  51  27  76  82  74   8  45  68
##  [79]  56  22  30  52  39  53   1  94  91  61  15  87  32  95  26  65  67  84  85  21  33  25
ru<- ru[order(ru)]
ru
##   [1]  1.536771  2.049110  2.674868  3.123521  4.114649  4.827495  5.822534  5.903874  6.987244  8.645009  9.320959  9.629041 10.457316 10.925103 12.860652
##  [16] 15.427447 15.572481 20.366357 20.778852 20.857380 21.123495 21.675698 23.171595 23.726180 23.848310 25.671108 26.782792 27.779676 30.176817 30.927743
##  [31] 31.236912 31.565060 31.661671 31.859265 32.578509 33.960568 36.092642 36.653669 36.900464 37.125076 37.690767 39.901974 41.499422 41.776966 41.845799
##  [46] 42.336720 42.826766 44.330766 44.351958 44.421337 44.535229 44.819617 50.624719 50.628892 54.702159 55.539444 57.439018 60.201699 61.515534 62.295121
##  [61] 65.183362 66.053051 66.387953 67.174919 67.894886 68.873884 69.391265 72.182660 72.994119 73.760295 74.111909 74.310003 74.752274 75.989280 76.080229
##  [76] 76.771348 76.898981 77.226256 79.600119 80.113802 80.367322 80.620398 80.669155 80.833172 82.128105 82.130361 82.832142 83.582061 85.309647 85.469616
##  [91] 86.677461 88.123850 89.996384 90.153031 91.736905 92.080373 95.683817 98.025970 98.318522 98.764508
plot(x = ru)

# 100 normally distributed numbers with a mean of 50 and sd of 10
nd <- rnorm(100, mean = 50, sd = 10)
nd
##   [1] 47.42441 57.33519 37.30232 38.79815 43.51726 51.25868 40.44545 35.39918 68.64166 45.78401 44.86283 42.35974 39.60551 35.83226 47.43445 57.63712 38.24823
##  [18] 56.57541 52.31066 50.72522 52.80729 45.35758 37.99853 30.48786 36.33766 58.00537 55.31346 51.04799 40.18687 46.24375 64.47137 41.86058 53.56293 43.03807
##  [35] 54.74270 57.36434 37.80610 50.51136 25.91483 42.59799 55.27973 51.27896 42.57362 41.72316 68.56577 50.80307 40.54199 45.68800 59.24867 47.08453 43.42005
##  [52] 53.13869 64.78146 59.46365 46.67372 67.01668 60.53596 45.37041 51.48604 49.47743 59.63229 35.41510 46.07102 31.94903 47.85520 50.07680 45.52613 44.97554
##  [69] 43.84678 47.13172 32.52161 29.17881 57.57086 49.55629 53.18765 48.44988 43.63571 27.91488 57.23042 39.91180 61.95241 40.39708 48.86655 54.19065 68.48433
##  [86] 48.26914 53.29390 46.51034 59.21309 49.63885 47.41289 66.84865 55.53836 61.54305 51.47196 39.44552 38.52188 76.26508 43.05119 53.61176
nd <- nd[order(nd)]
nd
##   [1] 25.91483 27.91488 29.17881 30.48786 31.94903 32.52161 35.39918 35.41510 35.83226 36.33766 37.30232 37.80610 37.99853 38.24823 38.52188 38.79815 39.44552
##  [18] 39.60551 39.91180 40.18687 40.39708 40.44545 40.54199 41.72316 41.86058 42.35974 42.57362 42.59799 43.03807 43.05119 43.42005 43.51726 43.63571 43.84678
##  [35] 44.86283 44.97554 45.35758 45.37041 45.52613 45.68800 45.78401 46.07102 46.24375 46.51034 46.67372 47.08453 47.13172 47.41289 47.42441 47.43445 47.85520
##  [52] 48.26914 48.44988 48.86655 49.47743 49.55629 49.63885 50.07680 50.51136 50.72522 50.80307 51.04799 51.25868 51.27896 51.47196 51.48604 52.31066 52.80729
##  [69] 53.13869 53.18765 53.29390 53.56293 53.61176 54.19065 54.74270 55.27973 55.31346 55.53836 56.57541 57.23042 57.33519 57.36434 57.57086 57.63712 58.00537
##  [86] 59.21309 59.24867 59.46365 59.63229 60.53596 61.54305 61.95241 64.47137 64.78146 66.84865 67.01668 68.48433 68.56577 68.64166 76.26508
plot(x = nd)

hist(x = nd)

hist(nd, breaks = 20, col = "darkgreen")

plot(x = density(nd))

boxplot(x = nd)

boxplot(x = nd, horizontal = T)

ggplot2

Lets be honest, the base plots are ugly! The ggplot2 package gives the user to create a better, more visually appealing plots. Additional packages such as ggbeeswarm and ggrepel also contain useful functions to add to the functionality of ggplot2.

library(ggplot2)
mp <- ggplot(xd, aes(x = x8, y = x9))
mp + geom_point()

mp + geom_point(aes(color = x3, shape = x3), size = 4)

mp + geom_line(size = 2)

mp + geom_line(aes(color = x3), size = 2)

mp + geom_smooth(method = "loess")

mp + geom_smooth(method = "lm")

xx <- data.frame(data = c(rnorm(50, mean = 40, sd = 10),
                          rnorm(50, mean = 60, sd = 5)),
                 group = factor(rep(1:2, each = 50)),
                 label = c("Label1", rep(NA, 49), "Label2", rep(NA, 49)))
mp <- ggplot(xx, aes(x = data, fill = group))
mp + geom_histogram(color = "black")

mp + geom_histogram(color = "black", position = "dodge")

mp1 <- mp + geom_histogram(color = "black") + facet_grid(group~.)
mp1

mp + geom_density(alpha = 0.5)

mp <- ggplot(xx, aes(x = group, y = data, fill = group))
mp + geom_boxplot(color = "black")

mp + geom_boxplot() + geom_point()

mp + geom_violin() + geom_boxplot(width = 0.1, fill = "white")

library(ggbeeswarm)
mp + geom_quasirandom()

mp + geom_quasirandom(aes(shape = group))

mp2 <- mp + geom_violin() + 
  geom_boxplot(width = 0.1, fill = "white") +
  geom_beeswarm(alpha = 0.5)
library(ggrepel)
mp2 + geom_text_repel(aes(label = label), nudge_x = 0.4)

library(ggpubr)
ggarrange(mp1, mp2, ncol = 2, widths = c(2,1),
          common.legend = T, legend = "bottom")

Statistics

# Prep data
lev_Loc  <- c("Saskatoon, Canada", "Jessore, Bangladesh", "Metaponto, Italy")
lev_Name <- c("ILL 618 AGL", "CDC Maxim AGL", "Laird AGL")
dd <- read_xlsx("data_r_tutorial.xlsx", sheet = "Data") %>%
  mutate(Location = factor(Location, levels = lev_Loc),
         Name = factor(Name, levels = lev_Name))
xx <- dd %>%
  group_by(Name, Location) %>%
  summarise(Mean_DTF = mean(DTF))
xx %>% spread(Location, Mean_DTF)
## # A tibble: 3 × 4
## # Groups:   Name [3]
##   Name          `Saskatoon, Canada` `Jessore, Bangladesh` `Metaponto, Italy`
##   <fct>                       <dbl>                 <dbl>              <dbl>
## 1 ILL 618 AGL                  47                    79.3               138.
## 2 CDC Maxim AGL                52.5                  86.7               134.
## 3 Laird AGL                    56.8                  76.8               137.
# Plot
mp1 <- ggplot(dd, aes(x = Location, y = DTF, color = Name, shape = Name)) +
  geom_point(size = 2, alpha = 0.7, position = position_dodge(width=0.5))
mp2 <- ggplot(xx, aes(x = Location, y = Mean_DTF, 
                      color = Name, group = Name, shape = Name)) +
  geom_point(size = 2.5, alpha = 0.7) + 
  geom_line(size = 1, alpha = 0.7) +
  theme(legend.position = "top")
ggarrange(mp1, mp2, ncol = 2, common.legend = T, legend = "top")

From first glace, it is clear there are differences between genotypes, locations, and genotype x environment (GxE) interactions. Now let’s do a few statistical tests.

summary(aov(DTF ~ Name * Location, data = dd))
##               Df Sum Sq Mean Sq  F value   Pr(>F)    
## Name           2     88      44    3.476   0.0395 *  
## Location       2  65863   32932 2598.336  < 2e-16 ***
## Name:Location  4    560     140   11.044 2.52e-06 ***
## Residuals     45    570      13                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

As expected, an ANOVA shows statistical significance for genotype (p-value = 0.0395), Location (p-value < 2e-16) and GxE interactions (p-value < 2.52e-06). However, all this tells us is that one genotype is different from the rest, one location is different from the others and that there is GxE interactions. If we want to be more specific, would need to do some multiple comparison tests.

If we only have two things to compare, we could do a t-test.

xx <- dd %>% 
  filter(Location %in% c("Saskatoon, Canada", "Jessore, Bangladesh")) %>%
  spread(Location, DTF)
t.test(x = xx$`Saskatoon, Canada`, y = xx$`Jessore, Bangladesh`)
## 
##  Welch Two Sample t-test
## 
## data:  xx$`Saskatoon, Canada` and xx$`Jessore, Bangladesh`
## t = -17.521, df = 32.701, p-value < 2.2e-16
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -32.18265 -25.48402
## sample estimates:
## mean of x mean of y 
##  52.11111  80.94444

DTF in Saskatoon, Canada is significantly different (p-value < 2.2e-16) from DTF in Jessore, Bangladesh.

xx <- dd %>% 
  filter(Name %in% c("ILL 618 AGL", "Laird AGL"),
         Location == "Metaponto, Italy") %>%
  spread(Name, DTF)
t.test(x = xx$`ILL 618 AGL`, y = xx$`Laird AGL`)
## 
##  Welch Two Sample t-test
## 
## data:  xx$`ILL 618 AGL` and xx$`Laird AGL`
## t = 0.38008, df = 8.0564, p-value = 0.7137
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -5.059739  7.059739
## sample estimates:
## mean of x mean of y 
##  137.8333  136.8333

DTF between ILL 618 AGL and Laird AGL are not significantly different (p-value = 0.7137) in Metaponto, Italy.

pch Plot

xx <- data.frame(x = rep(1:6, times = 5, length.out = 26),
                 y = rep(5:1, each = 6, length.out = 26),
                 pch = 0:25)
mp <- ggplot(xx, aes(x = x, y = y, shape = as.factor(pch))) +
  geom_point(color = "darkred", fill = "darkblue", size = 5) +
  geom_text(aes(label = pch), nudge_x = -0.25) +
  scale_shape_manual(values = xx$pch) +
  scale_x_continuous(breaks = 6:1) +
  scale_y_continuous(breaks = 6:1) +
  theme_void() +
  theme(legend.position = "none",
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        axis.text = element_blank(),
        axis.ticks = element_blank()) +
  labs(title = "Plot symbols in R (pch)",
       subtitle = "color = \"darkred\", fill = \"darkblue\"",
       x = NULL, y = NULL)
ggsave("pch.png", mp, width = 4.5, height = 3, bg = "white")

R Markdown

Tutorials on how to create an R markdown document like this one can be found here:


dblogr/


© Derek Michael Wright