Labile Metabolite Profiling in Human Blood Using Phosphorus NMR Spectroscopy

Discovery of Hypoxanthine and Inosine as Robust Biomarkers for Predicting the Preanalytical Quality of Human Plasma and Serum for Metabolomics

In cold human blood, the anomalous dynamics of adenosine triphosphate (ATP) result in the progressive accumulation of adenosine diphosphate (ADP), adenosine monophosphate (AMP), inosine monophosphate (IMP), inosine, and hypoxanthine. While the ATP, ADP, AMP, and IMP are confined to red blood cells (RBCs), inosine and hypoxanthine are excreted into plasma/serum. The plasma/serum levels of inosine and hypoxanthine depend on the temperature of blood and the plasma/serum contact time with the RBCs, and hence they represent robust biomarkers for evaluating the preanalytical quality of plasma/serum. These biomarkers are highly specific since they are generally absent or at very low levels in fresh plasma/serum and are highly sensitive since they are derived from ATP, one of the most abundant metabolites in blood. Further, whether blood was kept at room temperature or on ice could be predicted based on inosine levels. An analysis of >2000 plasma/serum samples processed for metabolomics-centric analyses showed alarmingly high levels of inosine and hypoxanthine. The results highlight the gravity of sample quality challenges with high risk of grossly inaccurate measurements and incorrect study outcomes. The discovery of these robust biomarkers provides new ways to address the longstanding and underappreciated preanalytical sample quality challenges in the blood metabolomics field.

MetDIT: Transforming and Analyzing Clinical Metabolomics Data with Convolutional Neural Networks

Clinical metabolomics is growing as an essential tool for precision medicine. However, classical machine learning algorithms struggle to comprehensively encode and analyze the metabolomics data due to their high dimensionality and complex intercorrelations. This article introduces a new method called MetDIT, designed to analyze intricate metabolomics data effectively using deep convolutional neural networks (CNN). MetDIT comprises two components: TransOmics and NetOmics. Since CNN models have difficulty in processing one-dimensional (1D) sequence data efficiently, we developed TransOmics, a framework that transforms sequence data into two-dimensional (2D) images while maintaining a one-to-one correspondence between the sequences and images. NetOmics, the second component, leverages a CNN architecture to extract more discriminative representations from the transformed samples. To overcome the overfitting due to the small sample size and class imbalance, we introduced a feature augmentation module (FAM) and a loss function to improve the model performance. Furthermore, we systematically optimized the model backbone and image resolution to balance the model parameters and computational costs. To demonstrate the performance of the proposed MetDIT, we conducted extensive experiments using three different clinical metabolomics data sets and achieved better classification performance than classical machine learning methods used in metabolomics, including Random Forest, SVM, XGBoost, and LightGBM. The source code is available at the GitHub repository at https://github.com/Li-OmicsLab/MetDIT, and the WebApp can be found at http://metdit.bioinformatics.vip/.