A new study reveals how USDA and ISU scientists leverage AI technology to tackle methane emissions in cattle, offering promising alternatives to harmful methane inhibitors like bromoform.
In a new study, published in the journal Animal Frontiers, researchers from the USDA Agricultural Research Service (ARS) and Iowa State University (ISU) have demonstrated the potential of generative artificial intelligence to speed up the identification of solutions to reduce enteric methane emissions from cattle — a significant contributor to both agricultural and overall U.S. greenhouse gas emissions.
“Developing solutions to address methane emissions from animal agriculture is a critical priority. Our scientists continue to use innovative and data-driven strategies to help cattle producers achieve emission reduction goals that will safeguard the environment and promote a more sustainable future for agriculture,” Simon Liu, an administrator at USDA-ARS, said in a news release.
Enteric methane emissions, produced during the digestion process in a cow’s rumen, account for approximately 33% of U.S. agriculture’s and 3% of total U.S. greenhouse gas emissions. Tackling these emissions is an essential step towards mitigating climate change and enhancing the sustainability of agriculture.
Focusing internally on the cow’s digestive system, the research team, leveraging computational models enhanced by AI, targeted bacteria involved in enteric fermentation. These bacteria produce methane as a byproduct, which is then expelled by cows through belching.
One promising substance identified is bromoform, a natural compound found in seaweed. Bromoform has demonstrated an ability to reduce bovine methane production by up to 98%.
However, due to its carcinogenic properties, its application in food-producing animals is limited. Consequently, the search for alternative, safe methane inhibitors is both urgent and challenging.
The innovative research approach utilized a combination of advanced molecular simulations and AI.
“We are using advanced molecular simulations and AI to identify novel methane inhibitors based on the properties of previously investigated inhibitors [like bromoform], but that are safe, scalable and have a large potential to inhibit methane emissions,” added co-author Matthew Beck, a research animal scientist with USDA-ARS during the study and now at Texas A&M University.
The researchers constructed large computational models from publicly available scientific data on cow rumens, allowing for AI algorithms to predict molecular behavior and identify promising candidates for further testing. The continuous loop of prediction and laboratory validation was facilitated by a machine learning model known as a graph neural network.
“Our graph neural network is a machine learning model, which learns the properties of molecules, including details of the atoms and the chemical bonds that hold them, while retaining useful information about the molecules’ properties to help us study how they are likely to behave in the cow’s stomach,” co-author Ratul Chowdhury, an assistant professor at ISU, said in the news release.
Through this iterative process, the scientists identified 15 molecules similar in function to bromoform that do not carry its toxic burden, marking significant progress in the quest for viable methane mitigation compounds.
Jacek Koziel, a research leader at USDA-ARS and a co-author of the study, emphasized the broader implications of this work.
“There are other promising strategies currently available to mitigate enteric methane emissions, but the available solutions are relatively limited. This is why combining AI with laboratory research, through iterative refinement, is a valuable scientific tool,” he said. “AI can fast-forward the research and accelerate these several pathways that animal nutritionists, researchers and companies can pursue to get us closer to a very ambitious goal of limiting greenhouse gas emissions and helping mitigate climate change.”
The detailed breakdown of computational and monetary costs per molecule provided by this study aims to guide future investments in similar research approaches.