Deep learning neural network prediction method improves proteome profiling of vascular sap of grapevines during Pierce's disease development

Plant secretome studies have shown the importance of plant defense proteins in the vascular system against pathogens. Studies on Pierces disease of grapevines caused by the xylem-limited bacteria Xylella fastidiosa (Xf) have detected proteins and pathways associated to its pathobiology. Despite the biological importance of the secreted proteins in the extracellular space to plant survival and development, proteome studies are scarce due to technical and technological challenges. Deep learning neural network prediction methods can provide powerful tools for improving proteome profiling by data-independent acquisition (DIA). We aimed to explore the potential of this strategy by combining it with in silico spectral library prediction tool, Prosit, to analyze the proteome of vascular leaf sap of grapevines with Pierces disease. The results demonstrate that the combination of DIA and Prosit increased the total number of identified proteins from 145 to 360 for grapevines and 18 to 90 for Xf. The new proteins increased the range of molecular weight, assisted on the identification of more exclusive peptides per protein, and increased the identification of low abundance proteins. These increases allowed the identification of new functional pathways associated with cellular responses to oxidative stress to be further investigated.