Recurring environmental extremes and disease events threaten large plantings of high-value perennial crops. Due to environmental and economic drivers, the demand for highest-quality, disease-free, true-to-type planting material in large volumes for vegetatively propagated high-value perennial crops is predicted to increase substantially in the next decades. This demand can be rapidly fulfilled through risk-free propagation of the plant material in a controlled environment using a tissue culture method called micropropagation. Currently, the commercial micropropagation process is carried out using simplistic, calendar-based scheduling software that lacks predictive functionality. NuPhY has developed a cloud-capable software platform, TreeTracker, with capacity to integrate such functionality. In order to enhance its utility, we propose to integrate a Design of Experiments module, which would depend on a Research Knowledge Base derived from published literature. In addition, NuPhY Inc. will evaluate the feasibility of using supervised machine learning in a high-throughput plant propagation process so that TreeTracker can detect key performance indicator data that can be parameterized for process improvements. The integration of supervised learning in TreeTracker is expected to provide insights regarding how best to optimize plant growth, maximize production output, and minimize costs for plant propagation.
