Spectronaut-nf

A nextflow pipeline to carry Spectronaut DIA analysis with its parallel execution attribute. Here, individual LC-MS/MS derived proteomic rawfiles are queued for Spectronaut search separately with the help of nextflow parallel execution attribute and the results are combined in the end to form an experiment wide output. This increases the Spectronaut DIA analysis search speed owing to the use of multiple High-Performance Computing (HPC) nodes with the help of batch schedulers (ex: SLURM).

Key Features

• 🚀 Parallel execution of Spectronaut searches across multiple HPC nodes
• 📈 Scalable processing of hundreds to thousands of DIA raw files
• ⚡ Reduced analysis time compared with standard Spectronaut execution
• 🔄 Fully resumable workflow using Nextflow caching
• 🖥️ Compatible with HPC schedulers such as SLURM
• 🧩 Configurable computational resources for each workflow stage
• 📊 Generates experiment-wide spectral libraries and DIA reports

Spectronaut-nf workflow

The pipeline consists of six main processes:

1. Library generation - Pulsar Stage 1
2. Generate QSP
3. Library generation - Pulsar Stage 3
4. Combine PSAR
5. DIA search
6. Combine/merge SNE

Installation

A detailed description on how to install the linux compatible SpectronautCMD.dll or the containerized image (*.sif) file is available in Deploying Spectronaut on an HPC system.

clone Spectronaut-nf repository

git clone https://github.com/BenCollinsLab/Spectronaut-nf
cd Spectronaut-nf

Prerequisite

The following software packages must be installed and accessible in your environment before running Spectronaut-nf.

Software	Minimum Version	Purpose
.NET	8.0.18	Required to execute the Spectronaut command-line application (SpectronautCMD.dll).
Nextflow	25.10.3	Workflow manager used to orchestrate and execute the pipeline processes.
SLURM	24.04.3	Optional but recommended HPC workload manager used to distribute and execute jobs in parallel across compute nodes.

Verify Installation

$ dotnet --info

Host:
  Version:      8.0.18
  Architecture: x64
  Commit:       ef853a7105
  RID:          rocky.8-x64
.
.
.

$ nextflow info
  Version: 25.10.3 build 10983
  Created: 22-01-2026 15:34 UTC (15:34 BST)
  System: Linux 4.18.0-553.64.1.el8_10.x86_64
  Runtime: Groovy 4.0.28 on Java HotSpot(TM) 64-Bit Server VM 22.0.2+9-70
  Encoding: UTF-8 (UTF-8)

$ sinfo --version
slurm 25.05.2

Note: Spectronaut-nf has been developed and tested using the software versions listed above. Earlier or later versions may work but have not been extensively validated.

Cluster Environment: On most HPC systems, these dependencies can be loaded through environment modules, for example:

module load dotnet/8.0.18
module load nextflow/25.10.3
module load slurm/25.05.2

Please consult your local HPC administrator if different module names or versions are used on your system.

Spectronaut-nf input variables

Users can set most of the input parameters required to run the Spectronaut-nf pipeline in params.yaml file. This file can be edited using vim or nano editor in linux/HPC platform.

DIA search inputs

Required Parameters

Define or modify all DIA search-related inputs in params.yaml:

Parameter	Description
JOB_NAME	Unique identifier/name for your Spectronaut search job
spec_bin	Absolute path to the Spectronaut command-line executable (SpectronautCMD.dll). If you use spec_bin, comment image_path.
image_path	Absolute path to the containerized Spectronaut image file (spectronaut.sif). If you use image_path, comment spec_bin.
license	Absolute path to your Spectronaut license file
rawfile_dir	Directory path containing all input LC-MS/MS raw data files (Default path is ./raw_d/)
FASTA	Protein sequence database file in bgsfasta format for peptide identification

Optional Parameters

Parameter	Description
sample_size	Number of raw files to randomly sample for library generation (if not specified, all files are used)
batch_size	Number of raw files to group together for each library generation process (i.e., task). Default: 1 (individual files processed separately via Pulsar)
excludePattern	Exclude raw files by name pattern (e.g., "pooled" will skip files containing "pooled")
COND_SETUP	TSV file specifying experimental condition/sample grouping metadata
REPORT	Spectronaut report template (.rs file) to customize output report columns and format
PROP_DIA	Optional custom Spectronaut search parameters file (.prop) for advanced search configuration
EXT_PSAR	External Spectronaut result file (.psar) to use as input for DIA search
max_files_for_ManageSNE	Maximum number of SNE result files threshold for manage SNE process execution. If the total number of raw files exceeds this cutoff, the pipeline will execute combineSNE process instead. Default: 150 (adjust based on available HPC/Cloud resources for large-scale SNE file merging)

HPC Configuration

Set the batch scheduler used in your HPC platform inside process.

executor = 'slurm'

This will be followed by setting up process-specific partition requirements such as CPUs, RAM and Duration available in the respective HPC partition.

Parameter	Description
*_queue	HPC queue/partition used for each task.
*_cpus	Number of CPU cores allocated per jtaskob.
*_memory	Memory allocated per task.
*_time	Maximum wall-clock time for task execution.

Note: Each process is executed by different modules in Spectronaut-nf i.e., SN_nf_pulsar (Process 1-3) , SN_nf_combine_psar (Process 4), SN_nf_dia (Process 5) and SN_nf_combine_sne (Process 6), each with configurable HPC resource requirements. Adjust the CPU, memory, and time limits based on your HPC cluster capabilities and data size.

Parallel Execution Control

• executor_queueSize - Maximum number of search tasks to execute in parallel on the HPC queue

Running Spectronaut-nf

nextflow -bg run /path/to/main.nf \
         -profile container \
         -params-file /path/to/params.yaml >> nextflow_cmd.log

Note: If users want to run Spectronaut in its native format (i.e., SpectronautCMD.dll) without containerizing it, either set -profile host or dont use it at all.

Resume Failed Runs

nextflow -bg run /path/to/main.nf \
         -profile container \
         -params-file /path/to/params.yaml \
         -resume >> nextflow_cmd.log

When users -resume in their command, Nextflow automatically resumes the search from previously completed tasks (Cached).

Output

Output Folder	Description
intermediates/	Task-specific outputs generated during the library creation stages (Pulsar Stage 1, Generate QSP, and Pulsar Stage 3).
out_lib/	Final library generation outputs, including the experiment-wide spectral library files (.psar and .kit) obtained by combining all task-specific .psar files.
out_dia/	Outputs from all DIA search tasks, including the final experiment-wide DIA search results. Users can use the final Spectronaut reports or SNE file for further downstream analysis.
logs/	Nextflow execution reports, timelines, traces, and other log files generated during the pipeline run.
work/	Nextflow working directory containing task-specific execution folders identified by unique hash keys.
tmp/	Temporary files and intermediate data generated during pipeline execution. These files can be safely removed after successful completion of the pipeline.

Example Directory Structure

results/
├── intermediates/
├── out_lib/
├── out_dia/
├── logs/
├── tmp/
└── work/

Testing and Continuous Integration

Spectronaut-nf uses GitHub Actions to validate the Nextflow code.

Local validation

Run:

nextflow lint .

This checks the Nextflow code for syntax and style issues.

Workflow testing

Linting does not guarantee that the pipeline executes successfully. Changes affecting workflow behaviour should also be tested with an appropriate dataset and execution profile.

Users can test Spectronaut-nf on their HPC system using the provided FASTA database and publicly available diaPASEF raw files from the PRIDE Archive. Since mass spectrometry raw files are large, they are not included directly in this repository.

Test Dataset

• FASTA database: Use the FASTA file provided in this repository.
• diaPASEF raw files: Download the raw files from the PRIDE Archive using the accession PXD078858.
• Spectronaut: Ensure that Spectronaut is installed and activated with a valid licence.
• HPC environment: Ensure that Nextflow and the required dependencies are available and that the pipeline is configured for your cluster's executor and resource limits.

Continuous Integration

The GitHub Actions workflow is defined in:

.github/workflows/ci.yml

It runs automated validation when the configured repository events occur.

Deploying Spectronaut on an HPC System

Spectronaut-nf requires an authorised installation of Spectronaut and a valid licence. Since Spectronaut is proprietary software, its installer and licence are not distributed with this repository.

Users can deploy Spectronaut on an HPC system using either a native installation or an Apptainer/Singularity container, depending on their cluster configuration and local policies.

Option 1: Native Spectronaut Installation

If Spectronaut is installed directly on the HPC system, configure spec_bin in params.yaml to point to the installed SpectronautCMD.dll.

Important: Do not copy only SpectronautCMD.dll from another machine. The command-line application may depend on additional files and libraries in the Spectronaut installation directory. The complete installation and its dependencies must remain accessible.

1. Obtain the Spectronaut Linux installer from an authorised Biognosys source.

2. Install Spectronaut according to the vendor's instructions. If administrator privileges are required, contact your HPC administrator.

3. Where supported, install Spectronaut in a persistent software directory accessible to the compute nodes. For example:

   /shared/apps/spectronaut/<version>/
   

   The actual directory structure depends on the installation method. Identify the location of SpectronautCMD.dll within the complete installation.

4. Ensure that the required .NET runtime is available in the job environment and that the installation and licence are accessible from the compute nodes.

5. Configure params.yaml with the appropriate paths:

   spec_bin: "/shared/apps/spectronaut/<version>/SpectronautCMD.dll"
   license: "/shared/licenses/spectronaut/<licence-file>"

   Replace these example paths with the actual locations on your HPC system.

Run the pipeline using the host execution profile, for example:

nextflow run main.nf \
    -profile host \
    -params-file params.yaml

Use the profile and parameter names defined in your local Nextflow configuration.

Option 2: Apptainer/Singularity Container

For HPC systems supporting Apptainer/Singularity, users can prepare a container image containing Spectronaut and its required runtime dependencies.

This approach packages the software environment into a single .sif image that can be stored in a shared directory and executed on compatible compute nodes.

Follow the instructions in the Spectronaut Container Setup Guide to:

1. Obtain the Spectronaut installer from an authorised source.
2. Build a Docker image containing Spectronaut and its required dependencies.
3. Convert the Docker image into an Apptainer/Singularity image.
4. Test the image and verify that the Spectronaut command-line application starts correctly.

After creating the image, copy it to a persistent location accessible to the compute nodes. For example:

/shared/containers/spectronaut/<version>/spectronaut.sif

Configure params.yaml to use the container:

image_path: "/shared/containers/spectronaut/<version>/spectronaut.sif"
license: "/shared/licenses/spectronaut/<licence-file>"

Then execute the pipeline using the container profile:

nextflow run main.nf \
    -profile container \
    -params-file params.yaml

Ensure that the container profile is configured correctly for your HPC environment and that the licence and input/output directories are accessible from within the container.

HPC Storage and Permissions

Regardless of the deployment method, consider the following requirements:

• Persistent storage: Store the Spectronaut installation or .sif image in a persistent software or project directory rather than temporary storage that may be automatically cleaned.
• Compute-node accessibility: Ensure that the software installation or container image, licence, input data, and output directories are accessible from the compute nodes executing the pipeline.
• File permissions: Configure appropriate read and execute permissions for the software installation or container image. Follow your institution's policies for licence-file permissions.
• Licensing: Obtain and activate Spectronaut according to the applicable licensing terms. Do not include licence files in the public repository or bake them into the container image.
• Security and distribution: Do not commit proprietary installers, licence files, or generated container images to this public GitHub repository.
• Testing: Test the installation or container before launching a large-scale analysis. Confirm that the runtime is available and that Spectronaut can execute a search using a valid licence.

Note: A message indicating that no valid licence was found may confirm that the command-line application starts, but it does not establish that the software is activated or ready for a search. Verify licensing separately before running the full pipeline.

Documentation and Community

• Changelog: CHANGELOG.md
• Report a bug or request a feature: GitHub Issues
• Contributing guidelines: CONTRIBUTING.md
• Testing and CI: See the Testing and Continuous Integration section above.
• Releases: GitHub Releases

Citation

Spectronaut-nf: A Nextflow Pipeline for Parallel Processing of DIA Data with Spectronaut bioRxiv, 2026