Utilizing Machine Learning with Phenotypic and Genotypic Data to enhance Effective Breeding in Agricultural and Horticultural Crops

Changing climate is driving frequent extreme weather events such as increased drought, and unpredictable seasonal precipitation, resulting in increased salinity, and shifts in pest pressures. Furthermore, changing climate has resulted in shifts in growing season periods and temperatures negatively impacting traditional crop production. Developing climate resilient crop varieties is a promising method to mitigate crop abiotic and biotic stresses while protecting the environment, global food systems, and human health. Recent advances in high-throughput genotyping technologies such as next-generation sequencing have enabled the effective identification of quantitative trait loci associated with such traits as stress resilience, environment variability, along with yield and quality components. However, accurate phenotypic assessment of large numbers of plants (e.g. breeding lines, or under variable field conditions) has been a major bottleneck as it is laborious and time-consuming, and inconsistent.

Accurate and precise assessment of a large number of germplasm lines can be achieved for some traits through collecting many phenotypic images and processing large volumes of images using Artificial Intelligence (AI)/Machine Learning (ML)/Deep Learning (DL) methods. Such AI/ML/DL-based high-throughput phenotyping platforms can significantly increase the ability to evaluate many genotypes objectively, repeatedly and effectively for a multitude of parameters. This in turn assists in quantitative trait locus (QTL) mapping and candidate gene identification, as a way to provide a greater understanding of the components of complex biological traits which will lead to breeding of superior varieties.

Research addressing the following issues will be welcomed for this Research Topic:

- Primary focus of this research topic is developing and/or utilizing high-throughput AI/ML/DL-based
phenotyping systems to effectively evaluate genetic variability.
- Novel approaches to evaluate large number of genotypes effectively and precisely using image
analysis techniques that will improve crop breeding.
- Processing phenotypic data by integrating it with existing genomic or other omics data through
machine learning approaches to identify and develop potential markers for selecting stress resilient
germplasm.
- Effective high-throughput phenotyping strategies that will assist breeding to meet future global
food, feed, and fiber demands, and will enhance ecosystem services in the face of climate change.

Keywords: Genetic Variability, High-throughput Phenotyping, Machine Learning, Deep Learning, Artificial Intelligence, Genotyping, Stress Resilience, Environmental Variability, Crop Yield and/or Quality, Crop Breeding QTL Mapping Horticultural and Agricultural Crops

Important note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.