Data type and structure

Data transformation or force coercion.

Now if, with everything you already know, you will better understand data transformation through examples, then, let’s make it little friend:

Attention attention!!!

The $ symbol we use it in R to “call” a column in our database

In this example it will be a little clearer. We have a database called Project and places that were worked were numbered, instead of having given them a name, what a problem! Thus, we will look at our first data and then we will see class type of the column Place, as we see in this code:

head(Project)

##   Place Sepal.Width Petal.Length Petal.Width Species
## 1     5         3.5          1.4         0.2  setosa
## 2     5         3.0          1.4         0.2  setosa
## 3     5         3.2          1.3         0.2  setosa
## 4     5         3.1          1.5         0.2  setosa
## 5     5         3.6          1.4         0.2  setosa
## 6     5         3.9          1.7         0.4  setosa

class(Project$Place)

## [1] "numeric"

How could you observe this column of data is identified by R as numeric, but we already know that they are not numbers, but rather that they are names of sites, for that reason, we have to transform them so that R does not mix them up and perhaps generate an error in our analysis. To do this, we will use as.character() function as we see in the next example

char <- as.character(Project$Place)
class(char)

## [1] "character"

And our intention was ready!

But… So I can only transform from numeric to character?

You can also perform other types of coercion, depending on what you need or as requested using commands that you will see in the next table:

How to use them? I tell you that they are used as we did with as.character()

Now, we will see how each of these structures works with useful examples.

Vectors

Vectors are the simplest data structure in R, as we saw in previous table, their dimension depends on the number of data it contains and it can only have one data type.

How can you tell that you have a vector and how can you create vectors?

You already know class() function that helps us to identify data type we have and sometimes data type structure; however, family of is() functions It is the most suitable for defining data structure as next code:

3

## [1] 3

is.vector(3)

## [1] TRUE

class(3)

## [1] "numeric"

How can you tell, when using both functions they give us different results, since class() function will indicate type of data and is() function. Being logical, it will tell us if it is (TRUE) or (FALSE), this type of response is known as Boolean.

Another example:

three <- 3
class(three)

## [1] "numeric"

is.vector(three)

## [1] TRUE

The way to create a vector is very simple; What’s more, all this time you have been watching it, we need <- operator and if you are going to add several data of same type you use c(). On the other hand, there is something very important that you must learn and that is way in which you are going to assign names to data structure. The important thing is that they mean something to you and that someone else can understand it.

More examples:

num <- c(1233,555,88,99,17)
class(num)

## [1] "numeric"

is.vector(num)

## [1] TRUE

nam <- c("Emma", "Charlotte", "Sophia", 
          "Aaron", "Caleb", "Wallace")
class(nam)

## [1] "character"

is.vector(nam)

## [1] TRUE

Now it’s your turn to practice, create different vectors, with different types of data, explore everything you can do, for example try adding numerical vectors and see what happens, as in next example:

ex1 <- c(112,667,99,56,47,12)
ex2 <- c(22,65,23.8,99,101,41)

sum <- ex1 + ex2
sum

## [1] 134.0 732.0 122.8 155.0 148.0  53.0

What happened? It remains for you to describe in the best way what happened. What if vectors do not have same number of data? Tell us!

Matrices y arrays

These structures are used in mathematics and statistics; What’s more, it is one of the most common and requested structures in many of packages that we usually use in Biology, and it is also one of reasons why we despair with R and want to hit keyboard like crazy.

Now, how to recognize and create matrices?

It’s so easy how to write matrix(), this function accepts two arguments, nrow (number of rows) and ncol (number of columns) to indicate number of rows and columns of matrix that you are going to generate, as in the next code

We can do this without specifying number of columns and rows. For this example you are going to generate a matrix from 1 to 12 as follows: 1:12

matrix(1:12)

##       [,1]
##  [1,]    1
##  [2,]    2
##  [3,]    3
##  [4,]    4
##  [5,]    5
##  [6,]    6
##  [7,]    7
##  [8,]    8
##  [9,]    9
## [10,]   10
## [11,]   11
## [12,]   12

Here R automatically shows us a matrix with 12 rows and a column, but suppose that you want 6 rows and 2 columns, for this we will enter as next code

matrix(1:12, nrow = 6, ncol = 2)

##      [,1] [,2]
## [1,]    1    7
## [2,]    2    8
## [3,]    3    9
## [4,]    4   10
## [5,]    5   11
## [6,]    6   12

Now create a matrix with numbers from 1 to 12 (1:12) with a nrow of 6 and ncol of 4, see what happens.

We can also create matrices by joining vectors with cbind() (join as columns) and rbind() (join as rows) functions as next form:

vec1 <- c(1:5)
vec2 <- c(10:15)
vec3 <- c(20:25)
vec4 <- c(35:40)

matrix1 <- cbind(vec1, vec2, vec3,vec4)
matrix1

##      vec1 vec2 vec3 vec4
## [1,]    1   10   20   35
## [2,]    2   11   21   36
## [3,]    3   12   22   37
## [4,]    4   13   23   38
## [5,]    5   14   24   39
## [6,]    1   15   25   40

matrix2<-rbind(vec1, vec2, vec3, vec4)
matrix2

##      [,1] [,2] [,3] [,4] [,5] [,6]
## vec1    1    2    3    4    5    1
## vec2   10   11   12   13   14   15
## vec3   20   21   22   23   24   25
## vec4   35   36   37   38   39   40

As you can see, these functions are very useful to us and allow us to carry out specific tasks.

With matrices you can also perform arithmetic operations.

matrix1 + 1

##      vec1 vec2 vec3 vec4
## [1,]    2   11   21   36
## [2,]    3   12   22   37
## [3,]    4   13   23   38
## [4,]    5   14   24   39
## [5,]    6   15   25   40
## [6,]    2   16   26   41

matrix1 * 2

##      vec1 vec2 vec3 vec4
## [1,]    2   20   40   70
## [2,]    4   22   42   72
## [3,]    6   24   44   74
## [4,]    8   26   46   76
## [5,]   10   28   48   78
## [6,]    2   30   50   80

matrix2 ^ 3

##       [,1]  [,2]  [,3]  [,4]  [,5]  [,6]
## vec1     1     8    27    64   125     1
## vec2  1000  1331  1728  2197  2744  3375
## vec3  8000  9261 10648 12167 13824 15625
## vec4 42875 46656 50653 54872 59319 64000

As you can see, arithmetic operation is performed for each data in the matrix

Now you will see data structure that is really most used by us in biology, data frames. But… Why is it the most used? It is the most used, because it is structure that best suits our needs; In addition, that is way in which we save our data in excel and then generally load file.csv

Generally rows in a data frame represent individuals or observations and columns represent attributes, traits or variables.

What does a data frame look like?

The iris data frame that we have been using, it is a clear sample of organization and data types that a data frame can contain.

But, how can I create a data frame in RStudio?

Quickly and easily, we will use data.frame() function, in which we can enter a series of vectors that will make up our data frame, in next example we will enter 4 vectors with 4 different data classes, this is one of advantages of a data frame, we can use different types of data.

Attention attention!!!

Remember that data frames require that variables be of same length. For this reason, we have to make sure that number of arguments passed to each vector are same

my_df <- data.frame(
  "integer" = 1:4, 
  "factor" = c("a", "b", "c", "d"), #  Note that characters must be between ""
  "number" = c(1.2, 3.4, 4.5, 5.6),
  "chain" = as.character(c("a", "b", "c", "d")))
  
my_df  

##   integer factor number chain
## 1       1      a    1.2     a
## 2       2      b    3.4     b
## 3       3      c    4.5     c
## 4       4      d    5.6     d

So we can see that our data frame has been created successfully, again we invite you to play with these functions and become more friends with them.

Finally we can use str() function to know structure of our data in a data frame, as we can see below

str(my_df)

## 'data.frame':    4 obs. of  4 variables:
##  $ integer: int  1 2 3 4
##  $ factor : chr  "a" "b" "c" "d"
##  $ number : num  1.2 3.4 4.5 5.6
##  $ chain  : chr  "a" "b" "c" "d"

Now you can observe structure of each of our data, 4 observations of 4 variables, in which we specify data type of each column or variable.