Collaborative Research: Trait-based prediction of microbial successional dynamics in thawing permafrost soils

Permafrost soils are perennially frozen soils that are common across arctic and boreal ecosystems. Soil microorganisms, largely dormant in intact permafrost, quickly become active when their environment thaws; they subsequently take advantage of undecayed material as a food source. As microbes grow, they consume organic forms of carbon and other key nutrients from the soil, and release inorganic forms of carbon like carbon dioxide. Since permafrost soils worldwide store approximately 50% of global soil carbon, these microbes have the potential to reshape carbon dynamics across our planet. Yet, the specifics of how communities of permafrost microbes respond to thaw and the factors that contribute to an organism’s success during this process remain largely undetermined. Existing tools for characterizing these microbes in the environment involve sequencing their DNA directly from soil samples, but these methods only give a snapshot of what is happening during thaw at a single moment in time. By developing new AI and machine-learning (ML) tools for analyzing microbial DNA sequences, combined with cutting-edge long-read DNA sequencing technology, this project will develop mathematical models that can take microbial DNA information and predict what happens to an entire community of microbes over time during permafrost thaw. In addition to advancing scientific knowledge, this project also brings the role of environmental microbes into focus for learners at all levels in New York City by (i) partnering with local undergraduate faculty and students to implement bioinformatics curricula and offer a summer undergraduate research experience for workforce development, and (ii) by partnering with a local community biology laboratory to develop a transdisciplinary outreach program that uses the visual arts to promote a broader public understanding of microbial life in soil and other environments. This project advances NSF’s priorities in Biotechnology and Artificial Intelligence. This project will develop a mechanistic, trait-based understanding of microbial succession during permafrost thaw, asking whether the growth strategies and environmental preferences of individual taxa in a community—predicted from genomic signals of trait evolution—can inform ecological models of bacterial and archaeal community dynamics in response to thawing of permafrost soils. First, this project will quantify the growth strategies of permafrost-dwelling microorganisms from metagenomes to understand organismal responses to soil thaw. Researchers will develop AI/ML methods for estimating the growth phenotypes and environmental preferences of microbial populations directly from genomic and metagenomic data to create a comprehensive classification of growth strategies in permafrost soils. Second, this project will integrate microbial growth strategies with ecological models of short-term community succession in thawing permafrost using high-resolution soil warming time-series experiments with longitudinal long-read metagenomic sequencing. Finally, this project will establish a pan-Arctic permafrost microbe growth parameter database to study long-term successional dynamics in microbial growth traits that will be made broadly available to earth systems modelers. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.