Wildland Urban Interface Exposure Toxicity in Cells, Animals, and Humans

Univ Of North Carolina Chapel Hill, Chapel Hill NC

Investigators

Abstract

PROJECT SUMMARY Cloud computing resources have the potential to propel NIH-support research efforts into territories that remain untapped, and to this day, hold immense promise to change our understanding of human health and the environment. This supplement proposes to develope the HeAlth Research eNablEd through Cloud ServiceS (HARNESS) program designed to implement advanced artificial intelligence / machine learning (AI/ML) methods using a high impact case study on wildfire research. This supplement is developed to support the parent ONES/R01 award, titled Wildland Urban Interface Exposure Toxicity in Cells, Animals, and Humans'. This parent award addresses the growing health impacts of wildfires, representing a threat to public health worldwide, growing in both intensity and prevalence year-by-year. The parent award specifically aimed to incorporate testing of the toxicological impacts of wildfire events occurring at the wildland urban interface (WUI), culminating in the burning of both anthropogenic and biogenic materials present in WUI region now facing increased risk of wildfire events. In the development of this parent award, a database has been cultivated and includes a wide range of exposure chemistries and toxicological responses (both at the -omics level and phenotypic lung functional level) associated with variable smoke exposures that can occur during wildfire events. With this database now developed and organized, we are poised as a team to further incorporate advanced AI/ML methodologies enhanced through cloud computing resources to better parse exposure-response profiles. This supplement specifically aims to leverage this starting resource to first develop the HARNESS workflow to be tested in the NIH Cloud Lab; then assess feasibility and cost-effectiveness of cloud services; and lastly to disseminate advanced AI/ML techniques and `lessons learned' from incorporating cloud computing into our research example through our online training resource, the inTelligence And Machine lEarning (TAME) Toolkit. This research effort will culminate into improved identification of chemical drivers, underlying disease mechanisms and resulting health outcomes of wildfire-associated exposure scenarios while enabling NIH researchers to more easily incorporate cloud-based resources into their everday research questions.