ZODIAC: database-independent molecular formula annotation using Gibbs sampling reveals unknown small molecules

Marcus Ludwig; Louis-Félix Nothias; Kai Dührkop; Irina Koester; Markus Fleischauer; Martin A. Hoffmann; Daniel Petras; Fernando Vargas; Mustafa Morsy; Lihini Aluwihare; Pieter C. Dorrestein; Sebastian Böcker

Throughout this paper, the entities of interest consist of signals detected by mass spectrometry where one or more MS/MS spectra have been recorded by the instrument. It is understood that not all of these signals correspond to compounds in the biological sample; but clearly, only those signals that do correspond are of interest for our analysis. It is also understood that we usually cannot ultimately decide whether a certain signal stems from the protonated molecule [M + H] + or, say, the protonated molecule with a water loss [M−H 2 O + H] + or an ammonia adduct [M + NH 3 + H] + . This is not a problem of our method but rather a general problem of mass spectrometry. For the sake of readability, we will nevertheless use the term compound instead of hypothetical compound, feature, adduct or ion. In contrast, our methods decide for each compound if it is protonated Diatoms. This dataset consists of solid phase (PPL, Agilent) extracts of the intra-and exometabolomes of a single diatom genus, a major group of marine microalgae. In total, ve culture samples of the species Pseudo-nitzschia subpacica (2×), Pseudo-nitzschia delicatissima (2×) and Pseudo-nitzschia multiseries (1×) were grown in culture from environmental isolates. Cultures included each diatom species and its associated microbiome resulting in a complex and diverse pool of metabolites. We expect the occurrence of compounds containing uncommon elements: Marine . CC-BY-NC-ND 4.0 International license author/funder. It is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/842740 doi: bioRxiv preprint microorganisms can contain halogenated organic compounds such as brominated molecules 36 ; additionally, the growth medium contained uncommon elements such as silicon and selenium. The samples were analyzed in data dependent acquisition (DDA) mode by UHPLC-MS/MS on a Q Exactive Orbitrap mass spectrometer in positive ionization mode. Metabolomics studies of marine algae are rare, and so, existing libraries are not expected to have many relevant entries, making this an interesting dataset for testing novel compound identication.