Harnessing Explainable AI to Identify Key Microbial Drivers of Reduced Ruminant Greenhouse Gas Emissions
Prof Chris Creevey, Queen's University Belfast; Dr Robert Atkinson, UoS
Scientific Background:
Microbes form stable communities by each taking on unique roles based on their genes, a process known as niche specialization. These communities can significantly impact their hosts, influencing health, efficiency, and environmental emissions such as methane in ruminants. This project aims to leverage explainable AI to analyse metagenomic microbiome data from ruminants, like cows, to understand how specific microbial communities can lead to lower emissions. The primary objectives will include identifying microbial communities associated with high and low emissions and using AI to determine which factors contribute to stable low-emission states.
​
Research Methodology:
The project will focus on deciphering the complex interactions within microbial communities and their ecological roles. By employing advanced AI techniques, the project will aim to unravel the intricate relationships that govern these communities. The project will benefit from the world-leading expertise in Queen’s University Belfast in microbial genomics, utilising our in-house Tier 2 High Performance Computing (HPC) facility and expertise in explainable AI from our project partners in Strathclyde University.
​
Training:
The successful candidate will have the opportunity to gain a deep understanding of explainable AI techniques and their applications in analysing microbial communities. They will develop skills in handling large datasets through advanced data science methods and gain expertise in bioinformatics, utilizing computational tools to study microbial genetics. The project will also provide a strong foundation in research methodology, including experiment design and result interpretation. Additionally, the successful candidate will be at the forefront of efforts to reduce the environmental impact of microbial communities on greenhouse gas emissions and sustainability. The interdisciplinary nature of the project will foster collaboration across fields such as microbiology, AI, and environmental science, equipping participants with valuable skills for careers in research, environmental management, and data science.
