A Comprehensive Database and Automated Analysis Tool for Stomatal Traits for Plants

Menglong Liu, Yin Wang
Institute of Ecology, College of Urban and Environmental Sciences, Key Laboratory for Earth Surface Processes of Ministry of Education, Peking University, Beijing, 100871 China

Stomatal traits are key indicators of plant responses to environmental stimuli, making the data availability and their accurate measurement essential for studies in plant physiology, ecology, and agricultural sciences. Existing methods often require burdensome manual collection and analysis, limiting data accessibility and slowing the pace of research. Here, we would like to introduce a project of a comprehensive online database of stomatal traits for various plant species, featuring high-resolution microscopic images of leaf stomata on both abaxial and adaxial epidermis (if possible) obtained through diverse methods (e.g. nail varnish-based surface impressions, epidermal separation and optical microscope imagery, direct scanning electron microscope imagery). Each image is accompanied by detailed metadata, including measurements of stomatal aperture and density, as well as information on the plants' growth conditions. To accelerate the analysis process, we developed a deep learning tool, named Stomaty demonstrated high efficiency in detecting and segmenting stomata, significantly outperforming manual methods while maintaining high accuracy. Our project aims to provide researchers with a valuable resource for studying stomatal physiology and environmental interactions, leveraging both extensive data and cutting-edge analytical tools.