Plant Systems

Discovery Engine for Extreme
Phenotypes and Field Investigation of Novel
Diversity (DEEPFIND)

Modeling the rules of life to program plant diversity for optimized traits.

Leads Anthony Studer, Michael Gore

Senior Personnel Meagan Lang, Amy Marshall-Colon, Joyce Van Eck, Steve Moose, Romit Roy Choudhury, Abraham Stroock, Duke Pauli, Robert Shepherd, Hakim Weatherspoon, Taryn Bauerle, Girish Chowdhary, Kelly Robbins

The DEEPFIND team will apply computational modeling to discover the Rules of Life, taking advantage of the tools and technologies developed by the CROPPS project.

Computational modeling is a powerful tool that allows us to better understand our world and make predictions. We will use models within an Internet of Living Things (IoLT) to perform intensive multiscale measurements of previously inaccessible traits—as well as of conventional ones—across training populations of genetically diverse programmable plants.

Programming extreme traits in individual genetic backgrounds will enable us to access deep biology and test model assumptions about interactions of biological pathways. We will then apply computational modeling to populations of programmed plants, which will allow us to discover and quantify a novel range of tunable traits across hundreds to thousands of distinct genetic backgrounds. The result will be greater understanding of the adaptability of a plant species to the various habitats and environments in which it lives, thereby expanding the adaptive potential of the species. In addition, we can iteratively refine our models and input data, updating and linking programs to alter a range of traits.

The DEEPFIND team seeks to develop programmable plants that can address the age-old adage that “you can only select for variation that you can observe.” Once we understand the relationship between a gene and a trait, we will be able to generate extreme, selectable trait variation that can ultimately improve genetic gains through the application of hybrid multi-scale models.

We will apply computational modeling to these populations of plants, which will allow us to do the following:

  • Discover and quantify a novel range of tunable traits across hundreds to thousands of distinct genetic backgrounds.
  • Increase understanding of the adaptability of a plant species across the various habitats and environments in which it lives, thereby expanding the adaptive potential of the population.
  • Generate extreme, selectable trait variation to inform hybrid multi-scale models and accelerate the improvement of genetic gains.

In our work, we will pursue the following objectives:

  • Objective 1: Program plants with extreme phenotypes and reporter circuits.
  • Objective 2: Investigate and quantify programmed plant responses to environmental stimuli.
  • Objective 3: Develop multi-scale models informed by deep biology and bio-physical parameters.
  • Objective 4: Predict plant response to the environment based on integrative multi-scale models.

Image Courtesy of TerraSentia