Fig. 3

Application of ML in understanding plant virus pathogenesis. ML enables early diagnosis of plant viral diseases at field level through analyzing hyperspectral images. Metagenomics study of diseased plant samples helps identification of related and unrelated viral genomes. ML can assist in the classification of these viral sequences which primes our understanding of virus evolution. Furthermore, ML-assisted bioinformatics tools have been developed to identify viral suppressors of RNA silencing (VSRs). ML can also guide us to predict the sub-cellular localization and even the structure of the viral proteins. Prediction of accurate structures of virus encoded proteins may help to identify inhibitors of these effector proteins. To understand the host response, several groups have performed transcriptome, proteome and metabolome of virus infected plants. ML can prime the accurate and fast analysis of these high throughput data to identify gene regulatory networks (GRN) and novel host factors involved in host-virus interplay. Characterization of these host factors in terms of sub-cellular localization and structure prediction will boost understanding of plant virus pathogenesis. ML may also assist plant virologists in genomic selection to identify elite virus resistant cultivars. This figure was created using BioRender (https://biorender.com/)