STAR: Modeling the phylogenetic architecture of biodiversity

About This Grant

Understanding how, when, and why organisms differ in their biological characteristics is key to interpreting the bewildering biodiversity of our planet. By providing insights into how lifeforms adapt, survive, and respond, this knowledge is crucial not only for ecology and evolutionary biology, but also for research in agriculture, biomedicine, and beyond. At the heart of this effort are statistical methods that enable data-driven investigations into the evolutionary and ecological mechanisms driving variation across individuals, populations, and species. Yet, progress is often hindered by analytical challenges involved in effectively linking variability observed in nature with the true phylogenetic history of organisms and their traits. These challenges are only expected to compound as the scale and scope of modern biological datasets continue to expand, demanding new, scalable solutions that consider the complexities of evolution and the shared ancestry of related species. Through an integrated analytics initiative, this project will synergize evolutionary biology and statistical learning to tackle unanswered questions in biodiversity research. Research-driven inquiry, infrastructure, and outreach will enrich statistical and biological learning, education, and mentoring in an EPSCoR state, and well beyond, through intensive education, training, and research missions driving intensive STEM workforce training. As a component of this work, this project will launch a new program to introduce STEM trainees to phylogenetics, biodiversity, and data science simultaneously. Early-career researchers will benefit from mentoring, training, networking, and exposure to STEM careers, as well as transdisciplinary skillsets in computational biology, data science, software development, and machine learning. This project will develop new statistical solutions for testing, predicting, and linking biological variation with the phylogenetic past when studying diverse traits, taxa, and questions. Specifically, this research seeks to transform the ways in which we evaluate evidence for hypotheses concerning the phylogenetic ancestry of species and their inherited characteristics. Researchers face a sea of phylogenetic trees of varying confidence, reliability, reproducibility, and applicability in the genomics era. There is therefore an urgent need to both understand the consequences of this uncertainty, and to create robust algorithms for modeling evolution in the context of complex, and often conflicting, phylogenetic information. This research will address both challenges through integrated studies that combine theoretical modeling, simulation, and empirical applications spanning diverse evolutionary and experimental conditions in comparative phylogenetics. By doing so, it aims to establish a framework for phylogenetic architectural analysis that is resilient to uncertainty and adaptable to the growing complexity of comparative data. This research therefore promises to advance evolutionary data science and build new frameworks for explicitly testing phylogenetic hypotheses about the historical processes underlying species-level variation in nature. Project deliverables will span a wealth of new analytical approaches that will be implemented in open-source, open-access software for the community to apply to most any similar trait, research question, and study. The research results and new approaches will benefit the research community and general public through advancements for comparative bio-logical studies with fundamental, biomedical, and agricultural applications. 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.