Humanity will need to engineer plants to withstand climate change and boost yield to feed an ever-expanding population. Moreover, the process of engineering better crops needs to be done efficiently, given the rapid rate of climate change. To improve efficiency, there must be clear paths to transfer engineering strategies between the wide range of agriculturally important species.

There have been several successful attempts to use AI to allow researchers to simulate genetic modifications on the computer, to enable more rapid plant genome engineering.

There have been recent developments in a field called Auto-Machine Learning (AutoML), which aims to democratise Artificial Intelligence by minimising the amount of expert intervention in the design of the AI system. Moreover, recent advances in Prior-data Fitted Network (PFN) models means that by training the AutoML on simulated datasets that show the kinds of relationships that the AI is expected to encounter, it is possible to apply AutoML to much smaller datasets.

You will explore whether AutoML and PFNs could be used to enhance our understanding of the structure of gene networks and how they evolve across flowering plants, including some agriculturally important crops like kale and oilseed. Moreover, you will use these AI techniques to accelerate plant engineering.

You will develop skills in AI, Bayesian statistics, horticulture, plant genetic engineering, and microscopy and have opportunities to collaborate with industrial partners and international researchers.

We are happy to consider individuals who are interested in changing disciplines. For instance, we are happy to support computer science, engineering or mathematics students who want to change their focus to ensure more sustainable food production under climate change. We are also equally happy to support students with a traditional biology background who want to learn AI, a discipline that will become ever-more important in our society.