Impact of matched samples on domain harmonization in cross-cohort NMR metabolomics.
6 affiliations
- 1Estonian Genome Centre, Institute of Genomics, University of Tartu, Tartu, Estonia. [email protected].
- 2Fondazione Istituto Italiano di Tecnologia, Analytical Chemistry Facility, Genoa, Italy. [email protected].
- 3Fondazione Istituto Italiano di Tecnologia, Genoa, Italy. [email protected].
- 4Fondazione Istituto Italiano di Tecnologia, Analytical Chemistry Facility, Genoa, Italy.
- 5Fondazione Istituto Italiano di Tecnologia, Analytical Chemistry Facility, Genoa, Italy. [email protected].
- 6Fondazione Istituto Italiano di Tecnologia, Genoa, Italy. [email protected].
Abstract
Introduction
Machine learning models trained on metabolomics data frequently show reduced performance when applied to independent cohorts because of domain shifts introduced by differences in analytical platforms, sample handling, storage conditions, and metabolite quantification procedures. Robust approaches for harmonizing metabolomics datasets across cohorts are therefore needed to improve model transferability and reproducibility.Objectives
To evaluate the impact of matched samples on domain harmonization and cross-cohort prediction performance in NMR metabolomics data, and to compare multiple harmonization strategies for improving model generalization across independently acquired datasets.Methods
A machine-learning framework was developed to classify ICD-10 code I50.9 (unspecified heart failure) using plasma NMR metabolomics data from the Estonian Biobank. Models were trained on a source dataset comprising 12,000 samples measured at Nightingale Health and evaluated on an independently measured destination dataset of 295 samples acquired on a different NMR platform. Harmonization approaches included within-domain normalization, Optimal Transport, Bridge mapping, and Empirical Bayes correction. The effect of matched samples was assessed using up to 108 duplicated biospecimens measured in both datasets. Additional validation was performed using independently generated human gestational diabetes and mouse diabetes metabolomics datasets from Metabolomics Workbench.Results
Domain harmonization generally improved cross-cohort prediction compared with unaligned data, particularly when matched samples were available. Predictive performance increased as the number of matched samples used for calibration increased, although gains varied across methods. Nevertheless, simple within-domain normalization consistently provided strong performance and often matched or exceeded more complex alignment approaches. External validation using fully unmatched human and mouse metabolomics datasets demonstrated that harmonization strategies improved classification performance relative to raw data, supporting the broader applicability of the framework.Conclusion
Matched samples substantially enhance domain harmonization and improve the transferability of metabolomics prediction models across analytical platforms and cohorts. However, robust normalization remains a strong baseline, and the benefits of more sophisticated alignment methods depend on the availability and representativeness of matched reference samples. These findings provide practical guidance for developing reproducible cross-cohort metabolomics machine-learning models while preserving biologically relevant signals.
Related
- Quality Parameters, Metabolomics Profile, and Biology Activities of Some Honey (Apis mellifera) Samples Obtained from Pretoria and Johannesburg, South Africa
- Comparative metabolomics of serum and serum extracellular vesicles of melanoma and triple-negative breast cancer utilizing biobank samples
- Leishmania infantum stage transition using untargeted metabolomics: an emphasis on reactive species impact
- Discoidin Domain Receptor 1 Expression in Colon Cancer: Roles and Prognosis Impact
- LLM-based cell type annotation harmonization across single-cell studies using GCTHarmony.
BibTeX
@article{ottas2026impact,
title = {{Impact of matched samples on domain harmonization in cross-cohort NMR metabolomics.}},
author = {A Ottas and L Goldoni and A Armirotti and S Decherchi},
journal = {Metabolomics : Official journal of the Metabolomic Society},
year = {2026},
month = oct,
doi = {10.1007/s11306-026-02543-0},
url = {https://europepmc.org/article/MED/42888323},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Impact of matched samples on domain harmonization in cross-cohort NMR metabolomics.
AU - A Ottas
AU - L Goldoni
AU - A Armirotti
AU - S Decherchi
PY - 2026
DA - 2026/10/17
JO - Metabolomics : Official journal of the Metabolomic Society
DO - 10.1007/s11306-026-02543-0
UR - https://europepmc.org/article/MED/42888323
AB - Introduction Machine learning models trained on metabolomics data frequently show reduced performance when applied to independent cohorts because of domain shifts introduced by differences in analytical platforms, sample handling, storage conditions, and metabolite quantification procedures. Robust approaches for harmonizing metabolomics datasets across cohorts are therefore needed to improve model transferability and reproducibility.Objectives To evaluate the impact of matched samples on domain harmonization and cross-cohort prediction performance in NMR metabolomics data, and to compare multiple harmonization strategies for improving model generalization across independently acquired datasets.Methods A machine-learning framework was developed to classify ICD-10 code I50.9 (unspecified heart failure) using plasma NMR metabolomics data from the Estonian Biobank. Models were trained on a source dataset comprising 12,000 samples measured at Nightingale Health and evaluated on an independently measured destination dataset of 295 samples acquired on a different NMR platform. Harmonization approaches included within-domain normalization, Optimal Transport, Bridge mapping, and Empirical Bayes correction. The effect of matched samples was assessed using up to 108 duplicated biospecimens measured in both datasets. Additional validation was performed using independently generated human gestational diabetes and mouse diabetes metabolomics datasets from Metabolomics Workbench.Results Domain harmonization generally improved cross-cohort prediction compared with unaligned data, particularly when matched samples were available. Predictive performance increased as the number of matched samples used for calibration increased, although gains varied across methods. Nevertheless, simple within-domain normalization consistently provided strong performance and often matched or exceeded more complex alignment approaches. External validation using fully unmatched human and mouse metabolomics datasets demonstrated that harmonization strategies improved classification performance relative to raw data, supporting the broader applicability of the framework.Conclusion Matched samples substantially enhance domain harmonization and improve the transferability of metabolomics prediction models across analytical platforms and cohorts. However, robust normalization remains a strong baseline, and the benefits of more sophisticated alignment methods depend on the availability and representativeness of matched reference samples. These findings provide practical guidance for developing reproducible cross-cohort metabolomics machine-learning models while preserving biologically relevant signals.
SN - 1573-3882
SN - 1573-3890
ER -