Fig. 1

Diagrammatic representation of the experimental procedure used to identify putative effectors from Myzus persicae genotype O, J and F, M. cerasi and Rhopalosiphum padi. (1) Aphids were dissected into biological replicas of heads and separately bodies (without nymphs). RNA was extracted and subjected to Illumina HiSeq sequencing. Following quality control (QC) and assembly, differential expression was performed to identify transcripts upregulated in head samples that encoded predicted signal peptides. These were categorised as putative effectors. (2) Aphid saliva was collected in artificial feeding chambers. The saliva was subjected to LC-MS/MS analysis. The resulting data was interrogated against the transcriptome assemblies in order to identify salivary secreted proteins. These were categorised as putative effectors. (3) Reciprocal best BLAST hit analysis was used to identity 1:1 ratio orthologues between M. persicae genotype O, J and F, M. cerasi, R. padi, Acyrthosiphon pisum and Aphis glycines. Clustering of the 1:1 ratio orthologous sequences was performed and where the resulting orthologous clusters contained a putative effector, they were subjected to DN/DS analysis. Clusters with a DN/DS value greater than 1 were identified as potentially under selection pressure. (4) Whole transcriptome clustering based on sequence similarity using BLAST and MCL, using the species listed above including Drosophila melanogaster, was used to identify clusters of putative effectors and those which maybe novel, termed pioneers in this study