Leveraging genomics and artificial intelligence to develop predictive pesticide risk assessment frameworks for wild pollinators

Bees are among the world's most environmentally and economically important group of insects, pollinating a remarkable diversity of flowering plants and playing a key role in the production of a wide range of food and commodity crops. However, while carrying out this ecosystem service bees can be exposed to a variety of potentially harmful toxins such as pesticides used in agriculture. Current bee pollinator pesticide risk assessment relies on experimental data collected for a handful of 'model' managed bee species such as the western honeybee. However, bees are a highly diverse group of insects comprising more than 20,000 known species. Thus, there is an urgent need to find ways to move beyond the use of data derived from a few managed model bee species as a proxy for wild bees in pesticide risk assessment. This project will leverage the dramatic increase in genomic information available for bees, in combination with recent advances in three-dimensional modelling of enzyme structures and enzyme expression, to develop new tools and predictive pipelines for bee pesticide risk assessment. The project will be conducted in partnership with a leading industrial partner who will provide additional funding for the project, and an excellent route to ensure the fundamental knowledge generated results in applied impact in terms of protecting bee pollinators.