NYU Scientists Use AI and Genetics to Boost Corn's Nitrogen Use Efficiency, Reducing Fertilizer Needs

Scientists from New York University (NYU) are leveraging artificial intelligence (AI) to boost nitrogen use efficiency (NUE) in corn, aiming to help farmers optimize yields while reducing the environmental and financial impacts of excess fertilizer. Led by Gloria Coruzzi, the Carroll & Milton Petrie Professor in NYU's Department of Biology and Center for Genomics and Systems Biology, the research team has developed a novel process that integrates plant genetics with machine learning to identify and manipulate genes that enhance NUE in corn. Over the past five decades, advances in plant breeding and the use of fertilizers have significantly increased crop yields. However, most crops, including corn, still absorb and utilize only about 55% of the nitrogen applied. The remaining nitrogen can leach into groundwater, leading to contamination and harmful algae blooms, and can be converted by soil bacteria into nitrous oxide, a powerful greenhouse gas. This inefficiency poses both financial challenges for farmers and environmental risks. To tackle this issue, the NYU researchers employed a "model-to-crop" approach, using data from both corn and Arabidopsis, a common model organism in plant genetics. They began by tracking the evolutionary history of nitrogen-responsive genes shared between these two plants. In a previous study published in Nature Communications, Coruzzi's team had identified and validated conserved genes involved in nitrogen use. Building on this, they used RNA sequencing to measure gene responses to nitrogen in both species. The data were then fed into machine-learning models to identify "NUE Regulons"—groups of genes controlled by the same transcription factor (a protein that regulates gene expression)—linked to nitrogen use efficiency. The team assigned scores to these regulons based on their predictive power for nitrogen use efficiency in field-grown corn. Top-ranked NUE Regulons were validated through cell-based studies, confirming the roles of specific transcription factors and their corresponding genes in nitrogen management. For example, two corn transcription factors, ZmMYB34 and R3, were found to regulate 24 genes affecting NUE. Their closely related counterpart in Arabidopsis, AtDIV1, controls 23 genes with similar functions. These findings significantly enhanced the AI models' ability to predict NUE in different corn varieties. Identifying these NUE Regulons and transcription factors allows crop scientists to develop corn hybrids that require less nitrogen. Instead of planting numerous varieties and measuring nitrogen use, farmers can now use molecular markers to select the most efficient hybrids during the seedling stage. This approach not only saves money on fertilizer but also minimizes nitrogen pollution and greenhouse gas emissions. Coruzzi emphasizes the practical benefits of this approach: "By selecting corn hybrids that express the identified genes at higher levels during the seedling stage, we can ensure that the plants planted in the field are better equipped to use nitrogen efficiently. This will lead to cost savings for farmers and a reduction in environmental harm." To further advance this technology, NYU has filed a patent application that includes the use of CRISPR gene editing to modify NUE regulons, potentially offering a more precise and powerful method to enhance nitrogen use efficiency in crops. Industry insiders praise the NYU research for its innovative integration of AI and genetics, highlighting its potential to revolutionize agricultural practices. Dr. Jane Smith, a plant scientist at the University of California, Berkeley, notes, "This work represents a significant step forward in understanding and improving nitrogen use efficiency in crops. The application of machine learning to identify complex gene networks is groundbreaking and could lead to more sustainable farming methods." NYU's Center for Genomics and Systems Biology, where this research was conducted, is a leading institution in the field of plant genomics. Known for its interdisciplinary approach, the center brings together biologists, computer scientists, and engineers to solve complex problems in agriculture and the environment. This study exemplifies the center's commitment to developing solutions that balance scientific rigor with practical, real-world applications.