quantmsdiann: a scalable SDRF-driven DIA-NN workflow for reanalysis of single-cell, spatial, and bulk proteomics datasets
Preprint
- 14 July 2026
- preprint
- Published by Springer Nature in Research Square
Abstract
Public proteomics archives now hold thousands of data-independent acquisition (DIA) datasets, but reusing them is difficult: each was processed with a different software configuration, and most lack standardized metadata. Here, we present quantmsdiann, an open-source Nextflow/nf-core workflow that runs DIA-NN in parallel across cloud and high-performance computing (HPC) infrastructure, guided by the experimental design declared in SDRF format. The workflow provides pinned container profiles and builds recipes for each supported DIA-NN version under BioContainers, resolving dependencies automatically, and reads all major vendor formats and the HUPO-PSI mzML standard. It exports harmonized quantification tables as MSstats input, in the Quantitative Proteomics eXchange (QPX) format, and a pmultiqc quality-control report. The parallel design reanalyzes a 2,300-run single-cell dataset in 2.2 hours on 300 HPC nodes. We performed multiple experiments and benchmarks of quantmsdiann on single-cell datasets; ProteoBench DIA-NN single-machine submissions or public datasets in ProteomeXchange. The benchmark against ProteoBench single-machine DIA-NN modules demonstrated no differences between DIA-NN single-machine runs and parallelization in quantmsdiann; while upgrading DIA-NN from 1.8.1 to a current release increased protein-group identifications by up to 17% in single-cell datasets. Remarkably, reanalysis of public DIA datasets with quantmsdiann and the latest version of DIA-NN always exceeds the originally deposited counts, recovering up to 59% more protein groups, with the largest gains on deposits processed with older or non-DIA-NN engines and smaller gains where a recent DIA-NN release was already used. quantmsdiann is a step toward scalable, reproducible reanalysis of the growing DIA data archive and a foundation for large-scale DIA meta-analysis and atlas building. It is available at https://github.com/bigbio/quantmsdiann.Keywords
All Related Versions
Funding Information
- Wellcome Trust (223745/Z/21/Z)
This publication has 35 references indexed in Scilit:
- pmultiqc: An Open-Source, Lightweight, and Metadata-Oriented QC Reporting Library for MS ProteomicsMolecular & Cellular Proteomics, 2026
- A framework for ultra-low-input spatial tissue proteomicsCell Systems, 2023
- Proteomic Dynamics of Breast Cancer Cell Lines Identifies Potential Therapeutic Protein TargetsMolecular & Cellular Proteomics, 2023
- MSstats Version 4.0: Statistical Analyses of Quantitative Mass Spectrometry-Based Proteomic Experiments with Chromatography-Based Quantification at ScaleJournal of Proteome Research, 2023
- Strategies for Increasing the Depth and Throughput of Protein Analysis by plexDIAJournal of Proteome Research, 2023
- The Current State of Single‐Cell Proteomics Data AnalysisCurrent Protocols, 2023
- Pan-cancer proteomic map of 949 human cell linesCancer Cell, 2022
- DIAproteomics: A Multifunctional Data Analysis Pipeline for Data-Independent Acquisition Proteomics and PeptidomicsJournal of Proteome Research, 2021
- Deciphering the signaling network of breast cancer improves drug sensitivity predictionCell Systems, 2021
- Scalable Data Analysis in Proteomics and Metabolomics Using BioContainers and Workflows EnginesProteomics, 2019