Computational effector prediction and population genomic analysis of wheat infecting Parastagonospora nodorum

Résumé : Parastagonospora nodorum is a major necrotrophic pathogen of wheat, causing substantial yield and economic loss in Australia (estimated $100M AUD per annum). Robust management of the disease requires an in-depth understanding of both the molecular and evolutionary dynamics involved in infection and pathogen populations. We sequenced and compared the whole genomes of 154 Western Australian P. nodorum isolates to identify the structure and diversity of the population, and to identify genes involved in pathogenicity. We found a single major diverse population, with several small near-clonal clusters restricted to specific times and locations. We observed high levels of mutation and a large accessory pangenome in the population, and identified several candidate proteins involved in virulence. To facilitate the identification of host-specific virulence proteins (effectors), we developed a pipeline for functional characterisation of proteins with an ML ranking model to prioritise candidate proteins. We also conducted a large scale study of fungal protein homology to identify remote homologues of effector proteins and further define structural families. By combining multiple levels of high-throughput experimental analysis and prediction methods, we were able to identify several genes involved in P. nodorum virulence. My project at INRAE will look at the plant host-specific RNA expression patterns observed in the pathogen Sclerotinia sclerotiorum, and what causes those changes in expression. The major project will involve a multi-omics experiment combining RNA-seq with ATAC-seq of S. sclerotiorum on multiple hosts. From this we will identify gene co-expression networks and ATAC-seq features associated with modules, which we will use to dissect the specific mechanisms involved in regulation of host-specific modules. Software will be developed to support this work.;