The use of mass defect vs retention time plot for unknown biomarker recognition, classification, and chemical formula prediction within LC-MS untargeted experiments

In LC-MS untargeted metabolomic experiments carried on high-resolution and high accuracy mass spectrometers, the biomarkers classification and annotation is a time-consuming, error-prone process. The injection of commercially available standards helps in the identification of common metabolites, but most of the signals come from unknowns/unexpected/non-commercially available compounds, and their identification is not straightforward. We propose to use the Retention Time and Mass Defect Ratio of the standards to create a 2 dimensional reference plot to assign a putative classification of the unknowns, and to narrow the number of their possible raw formulas. The results of a first application show that the plot allowed to classify and improve the assignment of chemical formulas to unknown biomarkers from Vitis Arizonica Texas berry skins