The apply function

The lesson introduces participants to the apply family of functions to efficiently perform operations across rows, columns and elements of tabular data.

Keywords

R, apply, apply family

Approximate time: 10 minutes

Learning Objectives

• Utilize the apply family of functions to a dataset

The apply family of functions

To obtain mean values for all samples we can use mean on each column individually, but there is also an easier way to go about it. Programming languages typically have a way to allow the execution of a single line of code or several lines of code multiple times, or in a “loop”. By default R is not very good at looping, hence the apply() family of functions are used for this purpose. This family includes several functions, each differing slightly on the inputs or outputs. For example, we can use apply() to execute some task on every element in a vector, every row/column in a dataframe, and so on.

Function	Description
apply()	Apply functions over array margins
by()	Apply a function to a data frame split by factors
eapply()	Apply a function over values in an environment
lapply()	Apply a function over a list or vector (returns list)
sapply()	Apply a function over a list or vector (returns vector)
mapply()	Apply a function to multiple list or vector arguments
rapply()	Recursively apply a function to a list
tapply()	Apply a function over a ragged array

We will be using apply() in the example below, but do take a moment on your own to explore the other related functions that are available.

The apply() function lets you apply another function (like mean(), sum(), etc.) across the rows or columns of a table of values. In the help pages, you’ll see this described as what “applying a function to margins” of a matrix.

In this context, margins correspond to:

• 1 - Rows
• 2 - Columns
• 1:2 - Each individual cell (row and column)

The syntax for the apply{} function is:

# DO NOT RUN
# apply() example syntax
apply(dataframe/matrix, margin, function_to_apply)

Let’s try this to obtain mean expression values for each sample in our RPKM matrix:

# Use apply mean across the columns of rpkm_ordered
apply(rpkm_ordered, 2, mean) 

  sample1   sample2   sample3   sample4   sample5   sample6   sample7   sample8 
10.266102 10.849759  9.452517 15.833872 15.590184 15.551529 15.522219 13.808281 
  sample9  sample10  sample11  sample12 
14.108399 10.743292 10.778318  9.754733