Getting to know a barley population's stomata: on how AI algorithms can improve stomatal density analysis

World hunger and climate changes are a challenge for this and future generations. With that in mind, barley, a cereal with a prominent position in animal fodder and stock food products, was the focus of this study regarding the improvement of its capacity to tolerate drought conditions. To this end, a study on a set of 208 different barley varieties was conducted to evaluate stomatal density in the population, along the different stages of its life cycle: the first leaf stage, the tillering, and growth of middle leaves, and the flag leaf heading stage. Within this project, a need for the development of informatics tools such as a stomatal counting algorithm was met. Namely, producing a high-throughput cloud-based artificial intelligence (AI) model with low error rates (2%), which is capable of measuring leaf areas and counting stomata of microscope images. The population presented stomatal densities comparable to or lower than those found in the literature, with the accessions Orria and SBCC145 to be subjected to further studies in the future due to their low stomatal density across all life stages. A low stomatal density causes a reduction in the water loss of the plant and the global prospect of less water availability for the plants in the future makes water conservation an advantageous feature. A genome-wide association study (GWAS) with the data collected should be the next point of action in developing this work further, to pick up on genetic correlations to this trait.