Light levels change throughout the day, are affected by climate and weather, and are filtered by the local environment. Switching between low and high levels of light over varying periods of time experienced by an organism in its environment shapes the tempo and mode of its light detection system. Plants must respond to dynamic environmental conditions and thus switch between efficient photosynthesis and photoprotection. Receptors on the plasma membrane perceive extracellular signals, such as photosynthetically-fixed sugars, are coupled to cytoplasmic G proteins to transduce information to cytoplasmic proteins and to amplify that signal to bring about changes like photosynthetic efficiency in both short (e.g. enzymatic reactions) and long (e.g. plant development) time scales. While G proteins have been shown to be important in regulating various aspects of stomata and photosynthesis, their role has yet to be fully understood. A regulator of G signaling (RGS) has been shown to sense sugars fixed in photosynthesis. Thus, we hypothesize that RGS mediates responses to dynamic light. The sequenced genomes within the grass family are the only genomes throughout Plantae known to lack RGS. By contrast, Setaria retains the RGS gene. Thus, the RGS gene from Setaria was expressed in rice to better understand the function of RGS. In this study, multiple transgenic events were grown to investigate their phenotypic response. We identified lines with altered stomatal patterning and rates of stomatal closure in response to changing light levels that will be used in future experiments.

Nitrogen inputs can be an important cost consideration for farmers in terms of economic profit as well as environmental impact. Elucidating genetic regions that are associated with plant phenotype response to nitrogen stress can help in facilitating breeding approaches that can mitigate these costs. A diverse population of 272 maize lines was planted at a field site in Champaign, IL in two consecutive years in reduced nitrogen conditions. 302 phenotypes were recorded including: seed ionomic content, root structural traits derived from 2-dimensional images as well as a 3-dimensional representation generated from x-ray computed tomography (XRT) scans, root traits extrapolated from mini-rhizotron systems, drone images across the growing season and end of season agronomic traits such as biomass and yield. BLUP models were fit to obtain estimates of single year genotypic values as well as across years values both of which took into account year specific spatial variation. Individual years and combined years BLUP values were used as response variables in genome wide association studies (GWAS) to identify loci significantly associated with each set of values. Significant associations were identified for all phenotype categories.

Stomata, microscopic pores on leaf surfaces, regulate the uptake of carbon dioxide and the simultaneous loss of water vapor by leaves. New image acquisition and analysis methods are allowing high-throughput phenotyping of stomatal patterning, which in turn have been applied to better understand the genetic basis of variation in certain species. However, it takes considerable data and effort to train the models, and their ability to accurately detect epidermal structures is constrained to morphologies found within the training data. This issue of context dependency, the inability to perform effectively in novel contexts, is the main hurdle preventing widespread adoption of machine learning in high-throughput phenotyping of intraspecific, interspecific, and environmental variation. Here we show the limited ability of a Mask-RCNN tool, which was previously trained and successfully applied to Zea mays, to analyze images from a closely related grass, Setaria viridis. We then demonstrate successful retraining of the tool to cope with the novel diversity presented by this new species. The stomatal complexes in optical tomography images of mature Setaria leaves were accurately identified by comparison to expert raters (R2 = 0.84). This study highlights the challenge of context dependency for widespread application of machine learning tools for phenotyping plant traits, even in closely related species. At the same time, it also provides a new tool that can be applied to leverage Setaria as a model C4 species, while also providing a roadmap for translation of a machine learning to analyze stomatal patterning in new plant species.