Every cell in a plant carries the same genetic blueprint, yet different cells use that blueprint in different ways, a process controlled by where and how genes begin to be read, or transcribed. This starting point, known as the transcription start site (TSS), acts like a switch that determines not only when a gene turns on but also what version of a protein it ultimately produces. Despite the central importance of these switches, scientists still lack a complete map of where they are located across the diversity of plant species, limiting our understanding of how plants regulate growth, development, and responses to their environment. This project will create new computational tools and experimental datasets to identify and characterize these critical genetic switches across plants ranging from algae to crops, with direct relevance to understanding and ultimately improving agriculturally important species such as corn. By advancing fundamental knowledge of how plant genomes are organized and regulated, this research supports the long-term national interest in strengthening agricultural science, training the next generation of scientists proficient in the use of rapidly advancing artificial intelligence (AI), genomics and molecular genetics technologies, and building a robust research infrastructure available to the broader scientific community.
This project brings together expertise in genetics, molecular biology, and computational biology to comprehensively map and functionally characterize TSSs in plants. The team will develop BioLearnTSS, a machine learning framework that predicts TSS locations and core promoter structure directly from DNA sequence, trained on existing data from flowering plants and then tested for generalizability using data from a conifer, a green alga, and a moss. The project will also investigate how the use of alternative TSSs shapes the proteome by altering splicing and translation efficiency, with a primary focus on maize. Using genome editing in maize and the model plant Arabidopsis, the team will determine how specific regulatory DNA elements within the core promoter influence TSS selection, gene expression, and protein production. The project is organized around three aims: (1) developing and validating machine learning models for plant TSS prediction across diverse lineages; (2) defining how alternative TSS usage affects splicing and translation; and (3) determining how core promoter elements shape TSS selection, mRNA accumulation, and splicing. Collectively, this work will generate foundational datasets, computational tools, and genome-edited resources that will be made publicly available to the broader research community, advancing both basic plant biology and agricultural innovation through biotechnology.
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.