Abstract
Among the cultivated crop species, the economically and culturally important grapevine plays host to the greatest number of distinctly characterized viruses. A critical component of the management and containment of these viral diseases in grapevine is both the identification of infected vines and the characterization of new pathogens. Next-generation high-throughput sequencing technologies, i.e., HTS technologies, have been widely adopted for their ability to quickly, broadly and directly characterize molecular sequences associated with potential pathogens. We empirically analyze the performance of HTS as a diagnostic tool in a phytosanitary context and make recommendations on its deployment for detecting known and novel viruses in grapevine. Three popular and widely used modalities for analyzing HTS data are characterized and compared using the standard diagnostic performance criteria of sensitivity (the true positive rate), specificity (the true negative rate) and analytical sensitivity (dilution series).