Deriving three one dimensional NMR spectra from a single experiment through machine learning

Nuclear Magnetic Resonance (NMR) spectroscopy is a powerful tool for analyzing complex mixtures due to its ability to manage matrix complexity, provide detailed molecular insights, and preserve sample integrity. In metabolomics, NMR enables the identification, quantification, and characterization of metabolites with minimal sample preparation, a broad dynamic detection range, and high reproducibility. Various NMR experiments, such as Nuclear Overhauser Effect SpectroscopY (NOESY), Carr-Purcell-Meiboom-Gill (CPMG), diffusion-edited, and J-resolved spectroscopy (JRES), offer complementary insights into biofluids like serum and plasma. However, acquiring multiple spectra for high-throughput applications can be resource-intensive and time-consuming. This study proposes a machine learning approach to predict CPMG, diffusion-edited, and JRES spectra directly from acquired NOESY spectra, leveraging serum samples as a case study to streamline analysis and improve efficiency in NMR-based metabolomics.