Mass Spectrometry-Based Targeted Lipidomics and Supervised Machine Learning Algorithms in Detecting Disease, Cultivar, and Treatment Biomarkers in Xylella fastidiosa subsp. pauca-Infected Olive Trees

• DOI: 10.3389/fpls.2022.833245
• Publication Date: 2022-04-22T06:20:36Z
• Authors: Valeria Scala; Manuel Salustri; Stefania Loreti; Nicoletta Pucci; Andrea Cacciotti; Giuseppe Tatulli; Marco Scortichini; Massimo Reverberi
• Keywords: Xylella fastidiosa; Olea

In 2013, Xylella fastidiosa (Xf) was detected for the first time in Apulia and, subsequently, recognized as causal agent of olive quick decline syndrome (OQDS). To contain disease, germplasm evaluated resistance to Xf, identifying cultivars with different susceptibility pathogen. Regarding this, resistant cultivar Leccino has generally a lower bacterial titer compared susceptible Ogliarola salentina. Among biomolecules, lipids could have pivotal role interaction Xf its host. grapevine Pierce's fatty acid molecules, diffusible signaling factors (DSFs), act regulators lifestyle and are crucial virulence. Other lipid compounds derived from oxidation, namely, oxylipins, can affect, vitro, biofilm formation subsp. pauca (Xfp) strain De Donno, that is, causing OQDS. this study, we combined high-performance liquid chromatography-mass spectrometry-MS-based targeted lipidomics supervised learning algorithms (random forest, support vector machine, neural networks) classify tree samples Salento. The dataset included either OQDS-positive or OQDS-negative trees belonging salentina treated not zinc-copper-citric biocomplex Dentamet®. We built classifiers using relative differences species able discriminate samples, (1) infected non-infected, (2) cultivars, (3) untreated Lipid entities emerging predictors thesis free acids (C16:1, C18:1, C18:2, C18:3); LOX-derived oxylipins 9- 13-HPOD/TrE; DOX-derived oxylipin 10-HPOME; diacylglyceride DAG36:4(18:1/18:3).