Plants are a vast source of complex chemical diversity that influence the outcomes of their interaction with insects, especially herbivores and herbivores’ predators and parasitoids. Plant domestication and modern breeding have profoundly reshaped plant chemical defense and interactions with other (insect) organisms. Recent years have seen a renewed interest in Crop Wild Relatives (CWRs) that is plants that were neither domesticated nor selected. Through comparisons of their retained genetic and metabolic diversity, CWRs offer opportunities to understand how we have shaped the genetic architecture and altered the chemical defenses of our crops.
In this seminar, I will provide an overview of how Machine Learning to identify candidate complex metabolites related to plant insect resistance and the challenges related to feature-rich high-resolution plant metabolomics. Further, I will describe a current project that combines computer vision, the use of genetic diversity and GWAS approaches to exemplify the use of CWRs to gain a better understanding of plant-insect interactions. Perspectives on future applications of Deep Learning in plant genomics and metabolomics will be presented to provide room for discussion.
For internal attendees
Symbiose seminars : https://www.cesgo.org/symbiose/seminars/machine-and-deep-learning-appli…