Open-Source Tools for High-Throughput Plant Phenotyping

Demand for food, fuel, and other plant products is expected to increase dramatically over the next century. At the same time, environmental considerations require that increases in agricultural output must occur using less water, land, fertilizer, and other inputs per unit of yield. One strategy to sustainably increase productivity is to develop new crops and cultivars that use resources more efficiently. While decreasing DNA sequencing costs has enabled rapid genetic screening of crop germplasm, only recently has the development of robotic imaging platforms and low-cost sensors led to major improvements in phenotyping throughput. Here we present PlantCV, an open-source framework for analyzing high-throughput plant phenotyping data. We demonstrate the utility of PlantCV and high-throughput phenotyping by analyzing the phenotypic diversity of a population of Camelina sativa natural accessions using the Bellwether Phenotyping Platform at the Donald Danforth Plant Science Center. C. sativa is an oilseed crop from the family Brassicaceae that is an emerging source of oil for fuel and is also being developed as a production platform for high-value compounds. Analysis of images taken daily for five weeks was used to measure natural diversity in above ground biomass, growth rates, days to flowering, and other traits. Additional analysis of seed phenotypes including yield per plant, seed size, and oil content was used to identify C. sativa accessions that could enhance breeding efforts. Although PlantCV was developed for the LemnaTec-based Bellwether Phenotyping Platform, we demonstrate that PlantCV can also be applied to low-cost phenotyping solutions and encourage community input in future development.