Evolving Comparative Genomics and Bioinformatics-Based Methods for the Prediction of Fungal Effectors

Bioinformatics analysis of whole-genome sequences has been fundamental in enabling powerful ‘reverse genetics’ approaches in molecular plant pathology, facilitating the functional validation of pathogenicity genes.  Thus genome-based bioinformatics provides a rational basis for prioritisation of the lab-based research efforts in this space.  In the last decade, “pathogenomics” has improved the understanding of gene content, protein properties and enzymatic functions for several pathogens.  Whole-genomes serve as a reference-point on which other data can be overlaid, such as transcriptomic, proteomic datasets or comparisons across multiple species.  Fungal phytopathogens possess a battery of molecules termed “effectors”, which disarm plant defences by either masking their presence or killing the host cell.  Identification of pathogenicity effectors is key in developing resistance strategies and bioinformatics is one of the primary methods of predicting the genes responsible for their production.  In particular, through comparison of gene content between species, comparative bioinformatics has highlighted patterns of gene presence or absence - which when correlated to phenotype can predict pathogenic function.  Comparative genomics has also furthered the study of mechanisms of genome evolution, by which pathogens may accelerate their rate of gene innovation and/or adaptation to host defences.    

We apply advanced bioinformatic techniques to enhance the understanding of pathogen evolution and host-interactions - with a view to informing concurrent molecular plant pathology research across a range of species. Our efforts begin with fundamental resources, making significant improvements to the reliability of in silico and RNA-seq-based predictions of gene annotations in Fungi.  Our work extends to the optimisation of bioinformatic techniques, such as the prediction of protein secretion - an important first step in effector prediction.  Broad-scale comparative genomics has been extensively used, with a focus on the prediction of host-range specificity and lateral gene-transfer, as a basis for the prediction of fungal pathogenicity genes.  Furthermore, narrow-scale comparative genomics has been utilised to examine gene presence and absence across isolates of single pathogen species with differential disease phenotypes, highlighting further effector candidates.  Our comparative genomics research also extends to furthering understanding of fungal genome mutagenesis mechanisms including mesosyntenic rearrangement and repeat-induced point mutation - which assists in the prediction of effector genes based on their genomic context.  The sum of these efforts is a multi-faceted approach to fungal effector prediction based on diverse sources of supporting evidence.