Revealing the Hidden Landscape of Public Metabolomics Data Reuse in MetaboLights

Abstract

Introduction: Metabolomics has historically lagged behind transcriptomics and proteomics in the scale and formalisation of public data reuse. Building on earlier manually assessed studies of metabolomics data reuse, this study provides an updated characterisation of confirmed reuse in MetaboLights. Objectives: This study aimed to characterise confirmed reuse of MetaboLights datasets in the published literature, quantifying its volume, reuse type, temporal trajectory, and dataset-level concentration. Methods: MetaboLights dataset accessions (MTBLS#) were used to retrieve candidate publications from five literature sources via a Python-based pipeline. Candidates were manually validated and classified by primary category and scientific domain; analytical modality was assigned from dataset metadata. Results: There were 274 confirmed reuse publications: Method or Tool Development (64.6%), Secondary Biological Analysis (16.8%), Data Integration or Meta-analysis (13.5%), and Educational Use or Protocol Illustration (5.1%). LC-MS was assigned to 70.1% of confirmed reuse publications; cumulative counts increased over time, and dataset-level reuse was highly asymmetric. Conclusion: Within the detected publications, MetaboLights datasets were predominantly reused for computational method development with secondary biological analysis being less frequent at present, underscoring the repository's value as a resource to obtain benchmarking datasets. Identifier-based searches may miss reuse activities when MTBLS identifiers are not consistently cited, reinforcing the community imperative for standardised dataset reporting and high-quality metadata deposition.