Operators

Last updated on 2025-07-05 | Edit this page

Overview

Questions

How do I perform operations, such as filtering, on channels?
What are the different kinds of operations I can perform on channels?
How do I combine operations?
How can I use a CSV file to process data into a Channel?

Objectives

Understand what Nextflow operators are.
Modify the contents/elements of a channel using operators.
Perform filtering and combining operations on a channel object.
Use the splitCsv operator to parse the contents of CSV file into a channel .

Operators

In the Channels episode we learnt how to create Nextflow channels to enable us to pass data and values around our workflow. If we want to modify the contents or behaviour of a channel, Nextflow provides methods called operators. We have previously used the view operator to view the contents of a channel. There are many more operator methods that can be applied to Nextflow channels that can be usefully separated into several groups:

Filtering operators: reduce the number of elements in a channel.
Transforming operators: transform the value/data in a channel.
Splitting operators: split items in a channel into smaller chunks.
Combining operators: join channels together.
Maths operators: apply simple math functions on channels.
Other: such as the view operator.

In this episode you will see examples, and get to use different types of operators.

Using Operators

To use an operator, the syntax is the channel name, followed by a dot . , followed by the operator name and brackets ().

GROOVY

channel_obj.<operator>()

view

The view operator prints the items emitted by a channel to the console appending a new line character to each item in the channel. We can also chain together the channel factory method .of and the operator .view() using the dot notation. Note: the view() operator doesn’t change the contents of the channel object.

Task 8.1

To make code more readable we can split the operators over several lines. The blank space between the operators is ignored and is solely for readability.

Show me the solution

GROOVY

//08_operators.nf

Channel
    .fromPath(params.school_info)
    .splitJson()
    .set{ pipe_school_info }

Closures

An optional closure {} parameter can be specified to customise how items are printed.

Briefly, a closure is a block of code that can be passed as an argument to a function. In this way you can define a chunk of code and then pass it around as if it were a string or an integer. By default the parameters for a closure are specified with the groovy keyword $it (‘it’ is for ‘item’).

Task 8.2

For example here we apply a closure to the queue channel, to separate the first two columns of the csv file as separate parameters and group all remaining columns into a single list of parameters.

Show me the solution

GROOVY

//08_operators.nf

Channel
    .fromPath(params.effects)    
    .splitCsv(header: false)
    .map { row -> [row[0], row[1], row[2..-1]] }
    .set{ pipe_effects }

Filtering operators

We can reduce the number of items in a channel by using filtering operators.

The filter operator allows you to get only the items emitted by a channel that satisfy a condition and discard all the others. The filtering condition can be specified by using either:

a regular expression
a literal value
a data type qualifier, e.g. Number (any integer,float …), String, Boolean
or any boolean statement.

Data type qualifier

Here we use the filter operator specifying the data type qualifier Number so that only numeric items are returned. The Number data type includes both integers and floating point numbers. We will then use the view operator to print the contents. To simplify the code we can chain multiple operators together, such as filter and view using a . .

The previous example could be rewritten like: The blank space between the operators is ignored and is used for readability.

Regular expression

We chain the .split() function, to extract the school ID from the input file name.

Task 8.3

Based on the example code in the 08_operators.nf file, explain the purpose of the split operator and intended output. Use the nextflow run 08_operators.nf to run the workflow and inspect the processs output using the .view() operator.

Show me the solution

Note we specify a regular expression .split("_|\\.") within the function in order to split the string based on the underscore “_” or punctuation “.” (whichever comes first) to derive an input variable, based on school ID. This is where generating file names dynamically as part of the workflow becomes relevant, as file names can be play an important role in managing the stream of data.

GROOVY

//08_operators.nf

workflow {
  .
  .
  .
    dataset =  GENERATE_DAT(ZipChannel_RDS) \
    | flatten \
    | map { file ->
    def key = file.name.toString().split('\\.')[0]
    def school_ID = file.name.toString().split("_|\\.")[0]
    return tuple(school_ID, key, file)}
  .
  .
  .
}

Boolean statement

A filtering condition can be defined by using a Boolean expression described by a closure {} and returning a boolean value.

Literal value

Finally, if we only want to include elements of a specific value we can specify a literal value.

Modifying the contents of a channel

If we want to modify the items in a channel, we can use transforming operators.

map

Applying a function to items in a channel

The map operator applies a function of your choosing to every item in a channel, and returns the items so obtained as a new channel. The function applied is called the mapping function and is expressed with a closure {}.

We can also use the map operator to transform each element into a tuple.

In the example below we use the map operator to transform a channel.

We can change the default name of the closure parameter keyword from it to a more meaningful name file using ->. When we have multiple parameters we can specify the keywords at the start of the closure, e.g. file, key ->.

Task 8.4

Inspect the code in the file 08_operators.nf explain the purpose of the map operator on the estimation_out.simulation_ch. How is it used to transform the contents into a tuple with the file and the file’s name? Write additional comments within the script. (Hint: Use the view operator to inspect the channel contents.)

Show me the solution

The simulation_ch output emits a tuple of elements as part of the simulation output from the ESTIMATION process. The map operator transforms the first of the elements indexed by [0] and uses a regular expression to split the character value on the first _ it encounters. Ex. it takes school123_period1 and returns school123, this allows us to generate a school identifier.

GROOVY

//08_operators.nf


workflow {
  .
  .
  .
    estimation_out.simulation_ch\
        | map { it -> [it[0].split('_|\\.')[0], it[1], it[2], it[3], it[4]]} 
  .
  .
  .   
}

Converting a list into multiple items

The flatten operator transforms a channel in such a way that every item in a list or tuple is flattened so that each single entry is emitted as a sole element by the resulting channel.

This is similar to the channel factory Channel.fromList.

Converting the contents of a channel to a single list item.

The reverse of the flatten operator is collect. The collect operator collects all the items emitted by a channel to a list and return the resulting object as a sole emission. This can be extremely useful when combining the results from the output of multiple processes, or a single process run multiple times.

The result of the collect operator is a value channel and can be used multiple times.

Grouping contents of a channel by a key.

The groupTuple operator collects tuples or lists of values by grouping together the channel elements that share the same key. Finally it emits a new tuple object for each distinct key collected.

If we know the number of items to be grouped we can use the groupTuple and size parameter. When the specified size is reached, the tuple is emitted. By default incomplete tuples (i.e. with less than size grouped items) are discarded (default).

This operator is useful to process altogether all elements for which there’s a common property or a grouping key.

Task 8.5

Inspect the code in the file 08_operators.nf explain the purpose of the groupTuple operator. How is it used to transform the contents into a tuple ? Write additional comments within the script. (Hint: Use the view operator to inspect the channel contents.)

Show me the solution

GROOVY

//08_operators.nf


workflow {
  .
  .
  .
    estimation_out.simulation_ch\
        | map { it -> [it[0].split('_|\\.')[0], it[1], it[2], it[3], it[4]]} \
        | combine(mapped_params, by: 0) \
        | transpose \
        | groupTuple(by: [1, 5], sort: true) 
  .
  .
  .   
}

Merging Channels

Combining operators allows you to merge channels together. This can be useful when you want to combine the output channels from multiple processes.

mix

The mix operator combines the items emitted by two (or more) channels into a single channel.

The items emitted by the resulting mixed channel may appear in any order, regardless of which source channel they came from. Thus, the following example it could be a possible result of the above example as well.

join

The join operator creates a channel that joins together the items emitted by two channels for which exists a matching key. The key is defined, by default, as the first element in each item emitted.

Maths operators

The maths operators allows you to apply simple math function on channels.

The maths operators are:

count
min
max
sum
toInteger

Counting items in a channel

The count operator creates a channel that emits a single item: a number that represents the total number of items emitted by the source channel. For example:

Splitting items in a channel

Sometimes you want to split the content of a individual item in a channel, like a file or string, into smaller chunks that can be processed by downstream operators or processes e.g. items stored in a CSV file.

Nextflow has a number of splitting operators that can achieve this:

splitCsv: The splitCsv operator allows you to parse text items emitted by a channel, that are formatted using the CSV format, and split them into records or group them into list of records with a specified length.
splitText: The splitText operator allows you to split multi-line strings or text file items, emitted by a source channel into chunks containing n lines, which will be emitted by the resulting channel.

splitCsv

The splitCsv operator allows you to parse text items emitted by a channel, that are formatted using the CSV format, and split them into records or group them into list of records with a specified length. This is useful when you want to use a sample sheet.

In the simplest case just apply the splitCsv operator to a channel emitting a CSV formatted text files or text entries. For example:

For the CSV file effects.csv.

We can use the splitCsv() operator to split the channel contaning a CSV file into three elements.

The above example shows hows the CSV file effects.csv is parsed and is split into three elements.

Accessing values

Values can be accessed by their positional indexes using the square brackets syntax[index]. So to access the first column you would use [0] as shown in the following example:

Column headers

When the CSV begins with a header line defining the column names, you can specify the parameter header: true which allows you to reference each value by its name, as shown in the following example:

Task 8.6

Inspect the 08_operators.nf, how is the params/effects.csv being parsed?

Show me the solution

Each row of the csv is read as a separate input. The closure using the map operator organises inputs by indexing the column order. The first two columns are stored as separate elements while the remaining columns are grouped into a list. The resulting input comprises of a tuple that involves 3 elements, two values and one list.

GROOVY

//08_operators.nf

Channel
    .fromPath(params.effects)    
    .splitCsv(header: false)
    .map { row -> [row[0], row[1], row[2..-1]] }
    .set{ pipe_effects }

Tab delimited files

If you want to split a tab delimited file or file separated by another character use the sep parameter of the split splitCsv operator.

More resources

See the operators documentation on the Nextflow web site.

Key Points

Nextflow operators are methods that allow you to modify, set or view channels.
Operators can be separated in to several groups; filtering , transforming , splitting , combining , forking and Maths operators.
To use an operator use the dot notation after the Channel object e.g. ESTIMATION.simulation_ch.view().
You can parse text items emitted by a channel, that are formatted using the CSV format, using the splitCsv operator.