Machine Learning-Enhanced Utilization of Plant Genetic Resources

Climate change profoundly affects our eco- and agro-ecosystems, and it is now a major concern and threat to world food and feed production security as well as biodiversity. Plant genetic resources (PGRs) serve as the reservoir of genetic adaptability, and the conservation and appropriate utilization of PGRs are keystones in adapting plant-based production systems to the effects of climate change and increasing food production to keep pace with the projected growth of the human population. Globally diverse PGRs in which appropriate genetic factors for use in climate change-affected production systems are well characterized and conserved and are, along with their associated knowledge, easily available to stakeholders will play an important role in supporting global food security and sustainable development. To achieve the above goals, comprehensive global-scale studies in which the interactions of genetic components of plant germplasm with detailed environmental and climate elements will be deliberate are required. In this chapter, we describe the sophisticated machine learning (ML) algorithms that are essential and fundamental tools for analyzing the large multi-dimensional datasets, which will be generated through the above studies. First, the main concepts and different classes of ML methods along with their properties will be discussed, and then the applications of ML-based high-throughput phenotyping and genotyping methods appropriate for sustainable use of PGRs in the breeding programs focused on developing climate-resilient germplasm will be described.