Abstract-LiquidChromatography-Mass Spectrometry (LC/MS) provides large datasets from which one needs to extract the relevant information. Since these data are made of non-negative mixtures of non-negative mass spectra, nonnegative matrix factorization (NMF) is well suited for its processing, but it has barely been used in LC/MS. Also, these data are very difficult to deal with since they are usually contaminated with non-Gaussian noise and the intensities vary on several orders of magnitude. In this article, we show the feasibility of the NMF approach on these data. We also propose an adaptation of one of the algorithms aiming at specifically dealing with LC/MS data. We finally perform experiments and compare standard NMF algorithms on both simulated data and an annotated LC/MS dataset. This lets us evaluate the influence of the noise model and the data model on the recovery of the sources.
Natural abundance deuterium 2D NMR spectroscopy in weakly ordering, polypeptide chiral liquid crystals is a powerful technique that enables determination of enantiotopic isotopic ratios ((2)H/(1)H)( i ) at the methylene groups of long-chain fatty acids. This technique has been used to study the bioconversion of linoleic acid to vernoleic acid with the objective of establishing the in-vivo site-specific fractionation of (2)H associated with this process. The fractionation pattern was investigated in Euphorbia lagascae and Vernonia galamensis, plants that use different enzyme systems to perform the Δ(12)-epoxidation: a cytochrome P450 monooxygenase in the former and a di-iron dioxygenase in the latter. The specific interest in this study was to ascertain whether different ((2)H/(1)H)( i ) isotopic ratios in substrate and product might reflect distinct features of the nature of the reaction centre. However, both the linoleate (substrate) samples and both vernoleate (product) samples isolated from the seed oils of the two plants had remarkably similar (2)H isotope profiles, with selection against (2)H in the positions around the Δ(12)-epoxidation site. This is interpreted as indicating that, despite differences in the form in which the activated Fe is presented and in the architecture of the active site, the ((2)H/(1)H)( i ) isotopic pattern is determined by features common to the reaction. It is suggested that the effects acting as the overall determinants of the final ((2)H/(1)H)( i ) distribution in the product are the encumbrance of the active site pocket and constraints to conformational readjustment during the linoleate to vernoleate transformation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.