Making proteomics reusable by design: interoperability standards for multi-omics and clinical translation

Proteomics has become a data-rich discipline, supported by high-throughput mass spectrometry (MS), public repositories, community standards, and increasingly scalable computational workflows. However, the public availability of proteomics datasets does not guarantee that they can be reanalyzed, compared with other studies, integrated with other omics layers, or used to support clinical interpretation.

This review addresses the standards, infrastructures, and analytical practices required to make proteomics reusable beyond the initial publication. We discuss the differences between data deposition, the FAIR (Findable, Accessible, Interoperable, and Reusable) principles, interoperability, and true reusability, and assess the roles of repositories, workflow-level reproducibility, multi-omics integration, and clinical phenotype harmonization. Common points of failure are highlighted: gaps in metadata, tool-specific outputs, misuse of missing values, batch-effect overcorrection, proteoform collapse, and premature claims of AI-readiness.

From data sharing as a final reporting obligation to interoperability as a design principle. Reusable-by-design proteomics requires that biospecimen context, metadata, spectral evidence, protein inference, quality control, workflow provenance, molecular specificity, and clinical meaning are all planned in concert from the outset of a study. The future impact of proteomics will depend not only on generating more data, but also on making those data independently interpretable, computable, auditable, and clinically reusable. For candidate medical tests, reusable-by-design should additionally anticipate metrological traceability, fit-for-purpose measurement uncertainty, and intended-use-specific analytical and clinical evidence so that clinically interpreted results remain comparable across laboratories, platforms, and time.