Applying comparative genomics to plant disease epidemiology

Abstract

Phylogenetic or genealogical interpretation of DNA sequence data from multiple genomic regions has become the gold standard for species delimitation and population genetics. Precise species concepts can inform quarantine decisions but are likely to reflect evolutionary events too far in the past to impact disease management. On the other hand, multilocus approaches at the population level can identify patterns of endemism or migration directly associated with episodes of disease, including host shifts and associated changes in determinants of pathogenicity and avirulence. We used the genome database of Magnaporthe grisea to frame a comparative, multilocus genomics approach from which we demonstrate a single origin for rice infecting genotypes with concomitant loss of sex in pandemic clonal lineages, and patterns of gain and loss of avirulence genes. In the Sclerotinia sclerotiorum pathosystem, we identified significant associations of multilocus haplotypes with specific pathogen populations in North America. Following the introduction of a new crop, endemic pathogen genotypes and newly evolved migrant genotypes caused novel, early-season symptoms.