FAIRaman: making clinical Raman spectroscopy data FAIR through biobanking and NeXus standards
- 1Istituti Clinici Scientifici Maugeri IRCCS, Via Maugeri 4, Pavia, Italy, 27100
- 2Department of Biomedical and Clinical Sciences, Università degli Studi di Milano, Via G.B. Grassi 74, Milan, Italy, 20157
Abstract
Raman spectroscopy is increasingly applied in biomedical and diagnostic research, generating spectral datasets that could support the development of robust artificial intelligence (AI) models for disease diagnosis, prognosis, and treatment guidance. However, most clinical Raman datasets remain difficult to reuse across laboratories because spectra are often stored separately from structured metadata describing the samples, their clinical context, experimental conditions, and data governance. This limits reproducibility, external validation, multicentre integration, and the development of transferable models. Here, we introduce FAIRaman, a FAIR-by-design framework and open-source software tool for the structured organisation of clinical Raman spectroscopy data. FAIRaman combines the hierarchical logic of the Investigation–Study–Assay (ISA) model with biobanking-derived descriptors for sample, donor, and processing-event annotation, and with NeXus/NXraman-inspired metadata for Raman acquisition parameters and spectral data. The resulting structure implemented in the open HDF5 file format. The software, developed in Python, converts native Renishaw WDF files and vendor-neutral text-based spectral files into FAIRaman-compliant HDF5 files through a graphical user interface (GUI), allowing users to combine batch-level metadata with sample-specific clinical data. We demonstrate the framework using a real Raman map of breast cancer tissue, and a Raman spectrum acquired from plasma-derived extracellular vesicles. These examples show that spectral data, acquisition metadata, and clinical sample descriptors can be stored within a single structured file and that the stored information can be easily retrieved using independent, freely available software tools.
Related
- Multidisciplinary Delphi consensus statement on minimal standards for clinical metadata and end points in microbiome studies
- Label-free Raman spectroscopy for cancer therapy response: from methodological framework to clinical translation.
- Rapid, Efficient Synthesis of Double-stranded DNA Standards for Quantitative PCR-Based Detection of HSV-2 using Polymerase Cycling Assembly.
- The Dual-Ribozyme RNA Ladder and RNA-Calibrated ssDNA Marker System for Inexpensive RNA Size Standards
- Differential histone tail citrullination by protein arginine deiminases observed via NMR spectroscopy
BibTeX
@misc{piccapietra2026fairaman,
title = {{FAIRaman: making clinical Raman spectroscopy data FAIR through biobanking and NeXus standards}},
author = {Davide Piccapietra and Mahin Vazifehdan and Arianna Bonizzi and Seyedeh Negin Sadat Madarshahi and Francesca Piccotti and Marta Truffi and Sara Albasini and Fabio Corsi and Carlo Francesco Morasso},
howpublished = {ChemRxiv},
year = {2026},
month = oct,
doi = {10.26434/chemrxiv.15005847/v2},
url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15005847/v2},
note = {CC BY},
}
RIS
TY - UNPB
TI - FAIRaman: making clinical Raman spectroscopy data FAIR through biobanking and NeXus standards
AU - Davide Piccapietra
AU - Mahin Vazifehdan
AU - Arianna Bonizzi
AU - Seyedeh Negin Sadat Madarshahi
AU - Francesca Piccotti
AU - Marta Truffi
AU - Sara Albasini
AU - Fabio Corsi
AU - Carlo Francesco Morasso
PY - 2026
DA - 2026/10/22
PB - ChemRxiv
DO - 10.26434/chemrxiv.15005847/v2
UR - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15005847/v2
AB - Raman spectroscopy is increasingly applied in biomedical and diagnostic research, generating spectral datasets that could support the development of robust artificial intelligence (AI) models for disease diagnosis, prognosis, and treatment guidance. However, most clinical Raman datasets remain difficult to reuse across laboratories because spectra are often stored separately from structured metadata describing the samples, their clinical context, experimental conditions, and data governance. This limits reproducibility, external validation, multicentre integration, and the development of transferable models. Here, we introduce FAIRaman, a FAIR-by-design framework and open-source software tool for the structured organisation of clinical Raman spectroscopy data. FAIRaman combines the hierarchical logic of the Investigation–Study–Assay (ISA) model with biobanking-derived descriptors for sample, donor, and processing-event annotation, and with NeXus/NXraman-inspired metadata for Raman acquisition parameters and spectral data. The resulting structure implemented in the open HDF5 file format. The software, developed in Python, converts native Renishaw WDF files and vendor-neutral text-based spectral files into FAIRaman-compliant HDF5 files through a graphical user interface (GUI), allowing users to combine batch-level metadata with sample-specific clinical data. We demonstrate the framework using a real Raman map of breast cancer tissue, and a Raman spectrum acquired from plasma-derived extracellular vesicles. These examples show that spectral data, acquisition metadata, and clinical sample descriptors can be stored within a single structured file and that the stored information can be easily retrieved using independent, freely available software tools.
ER -