Feature-Based Molecular Networking (FBMN) is a popular analysis approach for LC-MS/MS-based non-targeted metabolomics data. While processing LC-MS/MS data through FBMN is fairly streamlined, downstream data handling and statistical interrogation is often a key bottleneck. Especially, users new to statistical analysis struggle to effectively handle and analyze complex data matrices. In this protocol, we provide a comprehensive guide for the statistical analysis of FBMN results. We explain the data structure and principles of data clean-up and normalization, as well as uni- and multivariate statistical analysis of FBMN results. We provide explanations and code in two scripting languages (R and Python) as well as the QIIME2 framework for all protocol steps, from data clean-up to statistical analysis. Additionally, the protocol is accompanied by a web application with a graphical user interface (https://fbmn-statsguide.gnps2.org/), to lower the barrier of entry for new users. Together, the protocol, code, and web app provide a complete guide and toolbox for FBMN data integration, clean-up, and advanced statistical analysis, enabling new users to uncover molecular insights from their non-targeted metabolomics data. Our protocol is tailored for the seamless analysis of FBMN results from Global Natural Products Social Molecular Networking (GNPS and GNPS2) and can be adapted to other MS feature detection, annotation, and networking tools.

Background: C-peptide is produced in equimolar amounts with insulin from pancreatic beta cells, and thus is a fundamental biomarker for beta cell function. A non-invasive urinary C-peptide-to-creatinine ratio (UCPCR) has attracted attention as a biomarker for metabolic conditions. However, the UCPCR as an indicative risk predictor for prediabetes is still being investigated. Methods: We aimed to characterize UCPCRs in healthy people using American Diabetes Association (ADA) criteria and to evaluate their metabolic outcomes over time. A total of 1022 participants of the Biomarkers in Personalized Medicine cohort (BioPersMed) were screened for this study. Totals of 317 healthy with normal glucose metabolism, 87 prediabetic, and 43 diabetic subjects were included. Results: Prediabetic participants had a significantly higher UCPCR median value than healthy participants (p < 0.05). Dysglycaemia of healthy baseline participants was measured twice over 4.5 ± 0.9 years; 25% and 30% were detected with prediabetes during follow-ups, predicted by UCPCR both for the first (p < 0.05) and the second visit (p < 0.05), respectively. This is in good agreement with the negative predictive UCPCR value of 60.2% based on logistic regression. UCPCR levels were equal in both sexes. Conclusion: UCPCR measurements provide an indicative approach for metabolic risk, representing a potential use for prevention and monitoring of impaired glucose metabolism.