Introduction to Nextflow¶
Objectives
- Learn about the core features of Nextflow
- Learn Nextflow terminology
- Learn fundamental commands and options for executing pipelines
What is Nextflow?¶
Nextflow is a workflow orchestration engine and domain-specific language (DSL) that makes it easy to write data-intensive computational pipelines.
It is designed around the idea that the Linux platform is the lingua franca of data science. Linux provides many simple but powerful command-line and scripting tools that, when chained together, facilitate complex data manipulations.
Nextflow extends this approach, adding the ability to define complex program interactions and a high-level parallel computational environment based on the dataflow programming model.
Nextflow’s core features are:
- Pipeline portability and reproducibility
- Scalability of parallelization and deployment
- Integration of existing tools, systems, and industry standards
Whether you are working with genomics data or other large and complex data sets, Nextflow can help you to streamline your pipeline and improve your productivity.
Processes and Channels¶
In Nextflow, processes and channels are the fundamental building blocks of a pipeline.
A process is a unit of execution that represents a single computational step in a pipeline. It is defined as a block of code that typically performs one specific task. Each process will specify its input and outputs, as well as any directives and conditional statements required for its execution. Processes can be written in any language that can be executed from the command line, such as Bash, Python, Perl, or R.
Processes in are executed independently (i.e., they do not share a common writable state) as tasks. Multiple tasks can run in parallel, allowing for efficient utilization of computing resources. Nextflow is a top down pipeline manager and will automatically manage data dependencies between processes, ensuring that each process is executed only when its input data is available and all of its dependencies have been satisfied.
A channel is an asynchronous first-in, first-out (FIFO) queue that is used to join processes together. Channels allow data to passed between processes and can be used to manage data, parallelize tasks, and structure pipelines. Any process can define one or more channels as an input and output. Ultimately the pipeline execution flow itself, is implicitly defined by the channel declarations.
Importantly, processes can be parameterised to allow for flexibility in their behavior and to enable their reuse in and between pipelines. Parameters can be defined in the process declaration and can be passed to the process at runtime. Parameters can be used to specify the input and output files, as well as any other parameters required for the process to execute.
Execution abstraction¶
While a process defines what command or script is executed, the executor determines how and where the script is executed.
Nextflow provides an abstraction between the pipeline’s functional logic and the underlying execution system. This abstraction allows users to define a pipeline once and execute it on different computing platforms without having to modify the pipeline definition.
If not specified, Nextflow will execute locally. Executing locally is useful for pipeline development and testing purposes. However, for real-world computational pipelines, a high-performance computing (HPC) or cloud platform is often required.
Nextflow provides a variety of built-in execution options, such as local execution, HPC cluster execution, and cloud-based execution, and allows users to easily switch between these options using command-line arguments.
You can find a full list of supported executors as well as how to configure them in the Nextflow docs.
Nextflow CLI¶
Nextflow implements a declarative domain-specific language (DSL) that simplifies the writing of complex data analysis pipelines as an extension of a general-purpose programming language. As a concise DSL, Nextflow handles recurrent use cases while having the flexibility and power to handle corner cases.
Nextflow is an extension of the Groovy programming language which, in turn, is a super-set of the Java programming language. Groovy can be thought of as “Python for Java” and simplifies the code.
Nextflow provides a robust command line interface for the management and execution of pipelines. Nextflow can be used on any POSIX compatible system (Linux, OS X, etc). It requires Bash 3.2 (or later) and Java 11 (or later) to be installed.
Nextflow is distributed as a self-installing package and does not require any special installation procedure.
How to install Nextflow locally
- Download the executable package using either
wget -qO- https://get.nextflow.io | bash
orcurl -s https://get.nextflow.io | bash
- Make the binary executable on your system by running
chmod +x nextflow
- Move the nextflow file to a directory accessible by your
$PATH
variable, e.g,mv nextflow ~/bin/
Nextflow options and commands¶
Nextflow provides a robust command line interface for the management and execution of pipelines. The top-level interface consists of options and commands.
You can list Nextflow options and commands with the -h
option:
Usage: nextflow [options] COMMAND [arg...]
Options:
-C
Use the specified configuration file(s) overriding any defaults
-D
Set JVM properties
-bg
Execute nextflow in background
-c, -config
Add the specified file to configuration set
-config-ignore-includes
Disable the parsing of config includes
-d, -dockerize
Launch nextflow via Docker (experimental)
[truncated]
Commands:
clean Clean up project cache and work directories
clone Clone a project into a folder
config Print a project configuration
[truncated]
Options for a commands can also be viewed by appending the -help option to a Nextflow command.
For example, options for the the run command can be viewed:
Execute a pipeline project
Usage: run [options] Project name or repository url
Options:
-E
Exports all current system environment
Default: false
-ansi-log
Enable/disable ANSI console logging
-bucket-dir
Remote bucket where intermediate result files are stored
-cache
Enable/disable processes caching
-disable-jobs-cancellation
Prevent the cancellation of child jobs on execution termination
-dsl1
Execute the workflow using DSL1 syntax
Default: false
-dsl2
Execute the workflow using DSL2 syntax
Default: false
-dump-channels
Dump channels for debugging purpose
-dump-hashes
Dump task hash keys for debugging purpose
Default: false
[truncated]
Exercise
Find out which version of Nextflow you are using.
Managing your environment¶
You can use environment variables to control the Nextflow runtime and the underlying Java virtual machine. These variables can be exported before running a pipeline and will be interpreted by Nextflow.
For most users, Nextflow will work without setting any environment variables. However, to improve reproducibility and to optimise your resources, you will benefit from establishing some of these.
For example, for consistency, it is good practice to pin the version of Nextflow you are using with the NXF_VER
variable:
Exercise
Change the version of Nextflow you are using to 23.04.0
by exporting an environmental variable:
Solution
Export the singularity cache using the NXF_VER
environmental variable:
Check that the NXF_VER
has been applied:
You should see nextflow update and print the following:
Environmental variables on NeSI
The behaviour of Nextflow environmental variables won't work as expected if using a NeSI Nextflow module.
Similarly, if you are using a shared resource, you may also consider including paths to where software is stored and can be accessed using the NXF_SINGULARITY_CACHEDIR
or the NXF_CONDA_CACHEDIR
variables:
Exercise
Export the folder /nesi/nobackup/nesi02659/nextflow-workshop
as the folder where remote Singularity images are stored:
How to manage environmental variables
You may want to include these, or other environmental variables, in your .bashrc
file (or alternate) that is loaded when you log in so you don’t need to export variables every session.
A complete list of environmental variables can be found in the Nextflow docs.
Executing a pipeline¶
Nextflow seamlessly integrates with code repositories such as GitHub. This feature allows you to manage your project code and use public Nextflow pipelines quickly, consistently, and transparently.
The Nextflow pull
command will download a pipeline from a hosting platform into your global cache $HOME/.nextflow/assets
folder.
If you are pulling a project hosted in a remote code repository, you can specify its qualified name or the repository URL.
The qualified name is formed by two parts - the owner name and the repository name separated by a /
character. For example, if a Nextflow project bar
is hosted in a GitHub repository foo
at the address http://github.com/foo/bar
, it could be pulled using:
Or by using the complete URL:
Alternatively, the Nextflow clone
command can be used to download a pipeline into a local directory of your choice:
The Nextflow run
command is used to initiate the execution of a pipeline:
If you run
a pipeline, it will look for a local file with the pipeline name you’ve specified. If that file does not exist, it will look for a public repository with the same name on GitHub (unless otherwise specified). If it is found, Nextflow will automatically pull
the pipeline to your global cache and execute it.
Warning
Be aware of what is already in your current working directory where you launch your pipeline. If your current working directory contains nextflow configuration files you may encounter unexpected results.
Exercise
Execute the hello
pipeline directly from nextflow-io
GitHub repository.
Solution
Use the run
command to execute the nextflow-io/hello pipeline:
N E X T F L O W ~ version 23.04.0
Pulling nextflow-io/hello ...
downloaded from https://github.com/nextflow-io/hello.git
Launching `https://github.com/nextflow-io/hello` [silly_sax] DSL2 - revision: 1d71f857bb [master]
executor > local (4)
[e6/2132d2] process > sayHello (3) [100%] 4 of 4 ✔
Hola world!
Bonjour world!
Ciao world!
Hello world!
More information about the Nextflow run
command can be found in the Nextflow docs.
Executing a revision¶
When a Nextflow pipeline is created or updated using GitHub (or another code repository), a new revision is created. Each revision is identified by a unique number, which can be used to track changes made to the pipeline and to ensure that the same version of the pipeline is used consistently across different runs.
The Nextflow info
command can be used to view pipeline properties, such as the project name, repository, local path, main script, and revisions. The *
indicates which revision of the pipeline you have stickied and will be executed when using the run
command.
It is recommended that you use the revision flag every time you execute a pipeline to ensure that the version is correct.
To use a specific revision, you simply need to add it to the command line with the --revision
or -r
flag. For example, to run a pipeline with the v1.0
revision, you would use the following:
Nextflow automatically provides built-in support for version control using Git. With this, users can easily manage and track changes made to a pipeline over time. A revision can be a git branch
, tag
or commit SHA
number, and can be used interchangeably.
Exercise
Execute the hello
pipeline directly from the nextflow-io
GitHub using the v1.1
revision tag.
Solution
Use the nextflow run
command to execute the nextflow-io/hello
pipeline with the v1.1
revision tag:
N E X T F L O W ~ version 23.04.0
NOTE: Your local project version looks outdated - a different revision is available in the remote repository [3b355db864]
Nextflow DSL1 is no longer supported — Update your script to DSL2, or use Nextflow 22.10.x or earlier
Warning
The warning shown above is expected as the v1.1
pipeline revision was written using an older version of Nextflow that uses the depreciated echo
method.
As both Nextflow and pipelines are updated independently over time, pipelines and Nextflow functions can get out of sync. While most nf-core pipelines are now dsl2
(the current way of writing pipelines), some are still written in dsl1
and may require older version of Nextflow.
You can use an older version of nextflow on the fly by adding adding the environmental variable to the start of the run command
N E X T F L O W ~ version 22.10.0
NOTE: Your local project version looks outdated - a different revision is available in the remote repository [3b355db864]
Launching `https://github.com/nextflow-io/hello` [amazing_lovelace] DSL1 - revision: baba3959d7 [v1.1]
WARN: The use of `echo` method has been deprecated
executor > local (4)
[e6/cfda06] process > sayHello (4) [100%] 4 of 4 ✔
Bojour world! (version 1.1)
Hello world! (version 1.1)
Ciao world! (version 1.1)
Hola world! (version 1.1)
Nextflow log¶
It is important to keep a record of the commands you have run to generate your results. Nextflow helps with this by creating and storing metadata and logs about the run in hidden files and folders in your current directory (unless otherwise specified). This data can be used by Nextflow to generate reports. It can also be queried using the Nextflow log
command:
The log
command has multiple options to facilitate the queries and is especially useful while debugging a pipeline and inspecting execution metadata. You can view all of the possible log
options with -h
flag:
To query a specific execution you can use the RUN NAME
or a SESSION ID
:
To get more information, you can use the -f
option with named fields. For example:
There are many other fields you can query. You can view a full list of fields with the -l
option:
Exercise
Use the log
command to view with process
, hash
, and script
fields for your tasks from your most recent Nextflow execution.
Solution
Use the log
command to get a list of you recent executions:
TIMESTAMP DURATION RUN NAME STATUS REVISION ID SESSION ID COMMAND
2023-08-29 07:33:48 3.6s stupefied_bernard OK 1d71f857bb f9e18b71-d689-4589-be34-8cd98c1aab2e nextflow run nextflow-io/hello
Query the process, hash, and script using the -f
option for the most recent run:
Execution cache and resume¶
Task execution caching is an essential feature of modern pipeline managers. Accordingly, Nextflow provides an automated caching mechanism for every execution.
When using the Nextflow -resume
option, successfully completed tasks from previous executions are skipped and the previously cached results are used in downstream tasks.
Nextflow caching mechanism works by assigning a unique ID to each task. The task unique ID is generated as a 128-bit hash value composing the the complete file path, file size, and last modified timestamp. These ID's are used to create a separate execution directory where the tasks are executed and the outputs are stored. Nextflow will take care of the inputs and outputs in these folders for you.
A multi-step pipeline is required to demonstrate cache and resume. The christopher-hakkaart/nf-core-demo
pipeline was created with the nf-core create
command and has the same structure as nf-core pipelines. It is a toy example with 3 processes:
SAMPLESHEET_CHECK
- Executes a custom python script to check the input sample sheet is valid.
FASTQC
- Executes FastQC using the
.fastq.gz
files from the sample sheet as inputs.
- Executes FastQC using the
MULTIQC
- Executes MultiQC using the FastQC reports generated by the
FASTQC
process.
- Executes MultiQC using the FastQC reports generated by the
The christopher-hakkaartnf-core-demo
is a very small nf-core pipeline. It uses real data and bioinformatics software and requires additional configuration to run successfully.
To run this example you will need to include two profiles in your execution command. Profiles are sets of configuration options that can be accessed by Nextflow. Profiles will be explained in greater detail during the configuring nf-core pipelines section of the workshop.
To run this pipeline, both the test
profile and a software management profile (such as singularity
) are required:
This run requires 6GB of memory
If your session was spawned with less than 8GB of memory, above run will fail with the following error
The command line output will print something like this:
N E X T F L O W ~ version 23.04.0
Launching `https://github.com/christopher-hakkaart/nf-core-demo` [voluminous_kay] DSL2 - revision: 17521af3a8 [main]
------------------------------------------------------
,--./,-.
___ __ __ __ ___ /,-._.--~'
|\ | |__ __ / ` / \ |__) |__ } {
| \| | \__, \__/ | \ |___ \`-._,-`-,
`._,._,'
nf-core/demo v1.0dev-g17521af
------------------------------------------------------
Core Nextflow options
revision : main
runName : voluminous_kay
containerEngine : singularity
launchDir : /scale_wlg_persistent/filesets/home/chrishakk/session1
workDir : /scale_wlg_persistent/filesets/home/chrishakk/session1/work
projectDir : /home/chrishakk/.nextflow/assets/christopher-hakkaart/nf-core-demo
userName : chrishakk
profile : test,singularity
configFiles : /home/chrishakk/.nextflow/assets/christopher-hakkaart/nf-core-demo/nextflow.config
Input/output options
input : https://raw.githubusercontent.com/nf-core/test-datasets/viralrecon/samplesheet/samplesheet_test_illumina_amplicon.csv
outdir : results
Reference genome options
genome : R64-1-1
fasta : s3://ngi-igenomes/igenomes/Saccharomyces_cerevisiae/Ensembl/R64-1-1/Sequence/WholeGenomeFasta/genome.fa
Institutional config options
config_profile_name : Test profile
config_profile_description: Minimal test dataset to check pipeline function
Max job request options
max_cpus : 2
max_memory : 6.GB
max_time : 6.h
Generic options
tracedir : null/pipeline_info
!! Only displaying parameters that differ from the pipeline defaults !!
------------------------------------------------------
If you use nf-core/demo for your analysis please cite:
* The nf-core framework
https://doi.org/10.1038/s41587-020-0439-x
* Software dependencies
https://github.com/nf-core/demo/blob/master/CITATIONS.md
------------------------------------------------------
Downloading plugin nf-amazon@1.16.1
[f2/e5eb26] process > NFCORE_DEMO:DEMO:INPUT_CHECK:SAMPLESHEET_CHECK (samplesheet_test_illumina_amplicon.csv) [100%] 1 of 1 ✔
[bb/f98425] process > NFCORE_DEMO:DEMO:FASTQC (SAMPLE1_PE_T1) [100%] 4 of 4 ✔
[dd/728742] process > NFCORE_DEMO:DEMO:MULTIQC [100%] 1 of 1 ✔
-
Completed at: 29-Sep-2023 22:16:49
Duration : 2m 27s
CPU hours : (a few seconds)
Succeeded : 6
Executing this pipeline will create a work
directory and a results
directory with selected results files.
In the output above, the hexadecimal numbers, such as bb/f98425
, identify the unique task execution. These numbers are also the prefix of the work
directories where each task is executed.
You can inspect the files produced by a task by looking inside the work
directory and using these numbers to find the task-specific execution path:
The files that have been selected for publication in the results
folder can also be explored:
If you look inside the work
directory of a FASTQC
task, you will find the files that were staged and created when this task was executed:
The FASTQC
process runs four times, executing in a different work directories for each set of inputs. Therefore, in the previous example, the work directory [bb/f98425] represents just one of the four sets of input data that was processed.
To print all the relevant paths to the screen, use the -ansi-log
option can be used when executing your pipeline:
It's very likely you will execute a pipeline multiple times as you find the parameters that best suit your data. You can save a lot of spaces (and time) if you resume a pipeline from the last step that was completed successfully or unmodified.
By adding the -resume
option to your run
command you can use the cache rather than re-running successful tasks:
If you run
the christopher-hakkaart/nf-core-deme
pipeline again without making any changes you will see that the cache is used:
[truncated]
[5f/07e477] process > NFCORE_DEMO:DEMO:INPUT_CHECK:SAMPLESHEET_CHECK (samplesheet_test_illumina_amplicon.csv) [100%] 1 of 1, cached: 1 ✔
[b2/873706] process > NFCORE_DEMO:DEMO:FASTQC (SAMPLE2_PE_T1) [100%] 4 of 4, cached: 4 ✔
[ca/e8e0a8] process > NFCORE_DEMO:DEMO:MULTIQC [100%] 1 of 1, cached: 1 ✔
[truncated]
In practical terms, the pipeline is executed from the beginning. However, before launching the execution of a process, Nextflow uses the task unique ID to check if the work directory already exists and that it contains a valid command exit state with the expected output files. If this condition is satisfied, the task execution is skipped and previously computed results are used as the process results.
Notably, the -resume
functionality is very sensitive. Even touching a file in the work directory can invalidate the cache.
Exercise
Invalidate the cache by touching a .fastq.gz
file in a FASTQC
task work directory (you can use the touch
command). Execute the pipeline again with the -resume
option to show that the cache has been invalidated.
Solution
Execute the pipeline for the first time (if you have not already).
Use the task ID shown for the FASTQC
process and use it to find and touch
the sample1_R1.fastq.gz
file:
Execute the pipeline again with the -resume
command option:
You should that 2 of 4 tasks for FASTQC
and the MULTIQC
task were invalid and were executed again.
Why did this happen?
In this example, the cache of two FASTQC
tasks were invalid. The sample1_R1.fastq.gz
file is used by in the samplesheet twice. Thus, touching the symlink for this file and changing the date of last modification disrupted two task executions.
Your work directory can get very big very quickly (especially if you are using full sized datasets). It is good practise to clean
your work directory regularly. Rather than removing the work
folder with all of it's contents, the Nextflow clean
function allows you to selectively remove data associated with specific runs.
The -after
, -before
, and -but
options are all very useful to select specific runs to clean
. The -dry-run
option is also very useful to see which files will be removed if you were to -force
the clean
command.
Exercise
You Nextflow to clean
your work work
directory of staged files but keep your execution logs.
Listing and dropping cached pipelines¶
Over time, you might want to remove a stored pipelines. Nextflow also has functionality to help you to view and remove pipelines that have been pulled locally.
The Nextflow list
command prints the projects stored in your global cache folder ($HOME/.nextflow/assets
). These are the pipelines that were pulled when you executed either of the Nextflow pull
or run
commands:
If you want to remove a pipeline from your cache you can remove it using the Nextflow drop
command:
Exercise
View your cached pipelines with the Nextflow list
command and remove the nextflow-io/hello
pipeline with the drop
command.
Key points
- Nextflow is a pipeline orchestration engine and domain-specific language (DSL) that makes it easy to write data-intensive computational pipelines
- Environment variables can be used to control your Nextflow runtime and the underlying Java virtual machine
- Nextflow supports version control and has automatic integrations with online code repositories.
- Nextflow will cache your runs and they can be resumed with the
-resume
option - You can manage pipelines with Nextflow commands (e.g.,
pull
,clone
,list
, anddrop
)