Computational analysis of morphological changes in Lactiplantibacillus plantarum under acidic stress

Abstract

Cell shape and size often define characteristics of individual or communities of microorganisms in changing environments. Hence, characterizing cell morphology using computational image analysis can aid in the accurate identification of bacterial responses to these changes. Modifications in cell morphology of Lactiplantibacillus plantarum were determined in response to acidic stress, specifically during growth stage of the cells at pH 3.5 compared to pH 6.5. Consequently, we developed a computational method to sort, detect, analyze, and measure bacterial size in a single-species culture. We applied a deep learning methodology composed of object detection followed by image classification to measure the bacterial cell dimensions of the pre-identified cells. The results of our computational analysis show a significant change in cell morphology in response to alteration of environmental pH. Specifically, we found that the cell was dramatically elongated at low pH, while the width was not altered. Those changes could be attributed to modifications in membrane properties, for instance increased cell membrane fluidity in acidic pH. Integration of deep learning with microbial microscopic imaging is an advanced methodology for studying cellular structures. These trained models and scripts can be applied to other microbes and cells and are publicly available at: https://github.com/OraMoyal26/bacteria_dimensions/tree/main