oDIAFlow

An Open Nextflow pipeline for analyzing DIA-MS (Data-Independent Acquisition Mass Spectrometry) data.

Overview

oDIAFlow is a modular Nextflow pipeline for processing DIA proteomics data. The pipeline architecture follows the structure of quantms for future integration and compatibility. This repo is a sandbox for developing and testing DIA analysis workflows to ready them for inclusion in quantms.

Workflows

oDIAFlow currently implements two main workflows:

1. Empirical Library Workflow (DDA + DIA)

Generates a spectral library from DDA data, then uses it to analyze DIA data.

DDA mzML files
    ↓
SAGE Search (database search)
    ↓
EasyPQP ConvertSage (convert to pickle format)
    ↓
EasyPQP Library (build spectral library)
    ↓
OpenSwathAssayGenerator (generate transitions)
    ↓
OpenSwathDecoyGenerator (add decoys)
    ↓
    ┌─────────────────────────────────────┐
    │  DIA Analysis Pipeline              │
    └─────────────────────────────────────┘
    ↓
DIA mzML files → OpenSwathWorkflow (extract features + XICs)
    ↓
Merge OSW/OSWPQ (combine runs)
    ↓
Arycal (XIC-based alignment)
    ↓
PyProphet Alignment Scoring
    ↓
PyProphet (score → peptide inference → protein inference)
    ↓
Final Results (TSV)

2. In-Silico Library Workflow (Predicted Library + DIA)

Uses a predicted spectral library (e.g., from AlphaPeptDeep, DIA-NN) to analyze DIA data.

Transition TSV (predicted library)
    ↓
OpenSwathAssayGenerator (generate transitions)
    ↓
OpenSwathDecoyGenerator (add decoys)
    ↓
    ┌─────────────────────────────────────┐
    │  DIA Analysis Pipeline              │
    └─────────────────────────────────────┘
    ↓
DIA mzML files → OpenSwathWorkflow (extract features + XICs)
    ↓
Merge OSW/OSWPQ (combine runs)
    ↓
Arycal (XIC-based alignment)
    ↓
PyProphet Alignment Scoring
    ↓
PyProphet (score → peptide inference → protein inference)
    ↓
Final Results (TSV)

Usage

Empirical Library Workflow

nextflow run main.nf \
  --workflow empirical \
  --dda_glob "data/dda/*.mzML" \
  --dia_glob "data/dia/*.mzML" \
  --fasta "db/uniprot.fasta" \
  --irt_traml "lib/iRTassays.TraML" \
  --outdir results

In-Silico Library Workflow

nextflow run main.nf \
  --workflow insilico \
  --dia_glob "data/dia/*.mzML" \
  --transition_tsv "lib/predicted_library.tsv" \
  --irt_traml "lib/iRTassays.TraML" \
  --outdir results

Execution Environments

Local Execution

nextflow run main.nf -profile docker,local [options]

HPC Clusters (SLURM)

Generic SLURM cluster:

nextflow run main.nf -profile singularity,slurm [options]

Trillium (DigitalAlliance):

export SLURM_ACCOUNT=your-account
nextflow run main.nf -profile singularity,slurm,trillium [options]

Parameters

Required Parameters

Empirical workflow:

• --dda_glob: Path pattern to DDA mzML files (e.g., "data/dda/*.mzML")
• --dia_glob: Path pattern to DIA mzML files (e.g., "data/dia/*.mzML")
• --fasta: Path to FASTA database file

In-silico workflow:

• --dia_glob: Path pattern to DIA mzML files
• --transition_tsv: Path to predicted transition TSV file

Optional Parameters

• --workflow: Workflow type: 'empirical' (default) or 'insilico'
• --irt_traml: Path to iRT peptide TraML file
• --swath_windows: Path to SWATH window definition file
• --use_parquet: Use Parquet format for PyProphet (default: false)
• --outdir: Output directory (default: "results")
• --tools.arycal: Enable/disable Arycal alignment (default: true)

Tools Integrated

• Sage: Fast proteomics database search engine
• EasyPQP: Spectral library generation from search results
• OpenMS/OpenSWATH: Feature extraction and scoring
• Arycal: Chromatogram alignment
• PyProphet: Semi-supervised learning and statistical validation

Requirements

• Nextflow (version 25.10 or later)
• Docker or Singularity/Apptainer
• Container image: ghcr.io/openswath/openswath:dev

Configuration

Structure

conf/
├── base.config           # Resource labels (process_low, process_medium, etc.) 
├── modules.config        # Process-specific: ext.args, publishDir, containerOptions
├── profiles.config       # Execution profiles: docker, singularity, slurm, pbs, cloud
├── example_user.config   # Template for users to copy and customize
├── resources/
│   ├── local.config      # Resources for local/workstation execution
│   └── hpc.config        # Resources for HPC cluster execution
└── test/
    └── test.config       # Minimal test dataset configuration

nextflow.config           # Main config: params defaults + includes all sub-configs

Examples

# Local development with Docker
nextflow run main.nf -profile docker,local --dia_glob "data/*.mzML" ...

# HPC with Singularity and SLURM
nextflow run main.nf -profile singularity,slurm --dia_glob "data/*.mzML" ...

# Test run
nextflow run main.nf -profile docker,test

# Custom user config
nextflow run main.nf -profile docker,local -c my_analysis.config

Architecture

The pipeline follows a modular structure compatible with quantms:

oDIAFlow/
├── main.nf                    # Entry point
├── nextflow.config            # Configuration
├── workflows/                 # High-level workflows
│   ├── dia_empirical_library.nf
│   └── dia_insilico_library.nf
├── subworkflows/              # Reusable sub-workflows
│   └── local/
│       ├── pyprophet_osw/
│       └── pyprophet_parquet/
└── modules/                   # Individual process modules
    └── local/
        ├── sage/
        ├── easypqp/
        ├── openms/
        ├── pyprophet/
        └── arycal/