These are three examples of research projects in this research area:

1. Prescriptive cultivars: We explore the development of methods to identify and market prescriptive cultivars. Conceptually, a prescriptive cultivar is a variety that is developed for a specific agronomic production system, with a deeper understanding and utilization of the genotype, environment and phenotype and their interactions for variety development and placement.
Outcomes: established methods for development of new varieties that meet the specific requirements of different farmers, as per their production practices and environment, and customizes the product development and placement pipelines. The framework is adaptable and deployable in all row crop species and breeding system. Breeders can develop varieties that meets the specific requirement of different farmers, as per their production practices, and customizes the product development and placement pipeline establishing precision breeding.

2. Yield Prediction: We fuse high dimensional phenomics data with ML approaches to provide plant breeders needed tools for in-season seed yield prediction. These include short and long term studies including environment and climatic variables, which are integrated with genotype and molecular information for insight generation and actionable breeding outcomes.
Outcomes: We map complex relationships between phenotypic traits and seed yield, and prediction performance are assessed in different performance evaluation scenarios. Using ML methods and genetic algorithm techniques, we identify optimal wavebands for yield prediction. We develop prescriptive sensor package for high-throughput phenotyping deployment in breeding programs to minimize resource-intensive end-of-season phenotyping (e.g., seed yield harvest). The sensor enabled phenotyping and yield in-season yield prediction innovates the breeding pipeline.

3. Ground-based automated imaging: We develop and use ground-based imaging robots for saliency driven phenotyping with the intent to remove and/or minimize measurement variability in row crop phenotyping. These lightweight robots are equipped with digital cameras that acquire frequent measurements, and can work autonomously or with human control. We phenotype soybean accessions with this system, and in conjunction with ML-based organ feature extraction (compared to tedious human annotation) are advancing our work on the automation of phenotyping pipeline.
Outcomes: development of automated phenotyping systems that are equipped with versatile payloads (sensors) to alleviate phenotyping complexity.

Overall project outcomes and impact: We have widened genetic diversity, inputting useful genetic diversity, selecting more efficiently and accurately, and reducing the length of testing due to improved prediction models. Overall, these approaches help gain efficiencies in cultivar development and to enhance the rate of genetic gain as we favorably tweak the three parameters of the selection response (genetic variance, selection intensity, and repeatability).

Primary collaborators: Dr. Soumik Sarkar, Dr. Baskar Ganapathysubramanian.
Funding: Iowa Soybean Association, R. F. Baker Center for Plant Breeding, Monsanto Chair in Soybean Breeding, USDA-AFRI, Plant Sciences Institute.

Selected Publications:

• Parmley KA, RH Higgins, B Ganapathysubramanian, S Sarkar, AK Singh. 2019. Machine Learning Approaches for Prescriptive Plant Breeding. Scientific Reports, volume 9, Article number: 17132.
• Parmley K, K Nagasubramanian, S Sarkar, B Ganapathysubramanian, AK Singh. 2019. Development of Optimized Phenomic Predictors for Efficient Plant Breeding Decisions using Phenomics Assisted Selection in Soybean. Plant Phenomics, vol. 2019, Article ID 5809404.
• Singh AK, B Ganapathysubramanian, S Sarkar, A Singh. 2018. Deep learning for plant stress phenotyping: trends and future perspectives. Trends in Plant Science. 23(10): 883-898.
• Gao T, H Emadi, H Saha, J Zhang, A Lofquist, A Singh, B Ganapathysubramanian, S Sarkar, AK Singh, S Bhattacharya. 2018. A Novel Multirobot System for Plant Phenotyping. Robotics 7 (4): 61.
• Jubery T, J Shook, K Parmley, J Zhang, HS Naik, R Higgins, S Sarkar, A Singh, AK Singh, B Ganapathysubramanian. 2017. Deploying Fourier coefficients to unravel soybean canopy diversity. Front. Plant Sci. 7:2066.