Natural complex mixtures unequivocally defined in formulae difference space

Direct comparison of high-resolution mass spectrometry (HRMS) data acquired with different instrumentation or parameters remains difficult as the derived lists of molecular species via HRMS, even for the same sample, appear distinct. This inconsistency is the result of inherent inaccuracies caused by instrumental limitations and sample conditions. We propose a method that classifies HRMS data based on the differences in the number of elements between each pair of molecular formulae within the formulae list to preserve the essence of the given sample. The novel metric, Formulae Difference Chains Expected Length (FDCEL), allowed for comparing and classifying samples measured by different instruments. FDCEL metric was implemented for both spectrum quality control and for examination of samples of various nature. We also demonstrate a web application and a prototype for a uniform database for HRMS data serving as a benchmark for future biogeochemical applications.