Title: AI-enhanced gene-to-Earth system modeling
Abstract: Gene-to-Earth system modeling is one of the grant challenges of the DOE research mission. Advancing this topic can have broad impact by improving our understanding and application of omics-informed biological functional information in Earth science, reducing uncertainty in predicting coupled biogeochemical and hydrological cycles under changing hydroclimate, quantifying the biorecovery potential of critical minerals, and advancing bioenergy engineering. This presentation aims to discuss how we can integrate AI to advance gene-to-Earth system modeling, such as AI-based prediction of gene-informed microbial functional composition across the CONUS, AI-enhanced hybrid modeling to improve the prediction of soil biogeochemical cycles, as well as the AI emulator for filtering genotype-to-phenotype for bioenergy engineering.