Eves-van den Akker Group: AI-powered interrogation of natural variation in resistance and susceptibility to plant-parasitic nematodes
Supervisor:
Sebastian Eves-van den Akker
Importance of Research
Plant-parasitic nematodes threaten global food security. With at least one species able to parasitise every major food crop, they are estimated to cause over $100 billion in crop losses each year. The demand for new control strategies in line with the EU’s sustainability goals highlights significant knowledge gaps – we need to know more about how they cause disease in order to stop them. To generate a foundational data set we combined low-cost 3D printing of custom imaging machines with state-of-the-art deep-learning algorithms to make millions of measurements, of tens of thousands of parasites, infecting thousands of hosts, across hundreds of genotypes. This approach allowed us to reveal the previously unseen extent of host genetic control of parasite development, define new and unexpected physiological limits of the interaction, and discover novel fundamental, host-genotype-independent, features of parasite biology.
Project Summary
This new project will leverage these exceptionally rich data sets of nematode and plant traits from hundreds of genetic backgrounds to understand the plant genes which determine the outcome of infection by nematodes. The project will combine statistical genetics and the latest generative AI techniques, to
address the following main aims:
1. To determine the causal genes underlying natural variation in resistance/susceptibility to plant-parasitic nematodes
2. To determine the efficacy of using generative AI to predict cross-kingdom interactions from genomic data.
What the successful applicant will do?
The exact nature of the experiments will of course be tailored to the experience and/or interests of the successful applicant. However, experiments which address the above aims include Genome Wide Association Studies (GWAS), genetic mapping and phenotyping, training and testing generative AI models, and tDNA mutational analyses. There is an opportunity for the successful applicant to be mentored in the development of their own ideas as part of the project.
Training
No prior experience is required. The successful candidate will receive hands-on training in all required techniques and approaches by experienced members of the group and have unlimited access to training courses provided by the University of Cambridge. If you would like to explore an important and interesting biological question while building/refining hands-on skills in AI then this is the project for you!
References
Sebastian Eves-van den Akker, Plant–nematode interactions, Current Opinion in Plant Biology,
Volume 62, 2021, 102035, ISSN 1369-5266, https://doi.org/10.1016/j.pbi.2021.102035.