Data-driven analysis for the evaluation of cortical mechanics of non-adherent cells.

Atomic Force Microscopy (AFM) analysis of single cells, especially nonadherent, is inherently slow and analysis-heavy. To address the inherent difficulty of measuring individual cells, and to scale up toward a large number of cells, we take a two-fold approach; first, we introduce an easy-to-fabricate reusable poly(dimethylsiloxane)-based array that consists of micron-sized traps for single-cell trapping, second, we apply a deep-learning method directly on the extracted curves to facilitate and automate the analysis. Our approach is validated using suspended cells, and by applying a small compression with a tipless cantilever AFM probe, we investigate the effect of various cytoskeletal drugs on their deformability. We then apply deep learning models to extract the elasticity of the cell directly from the raw data (with a Coefficient of Determination of 0.47) as well as for binary (with an Area Under the Curve score of 0.91) and multi-class classification (with accuracy scores exceeding 0.9 for each drug). Overall, the versatility to fabricate the microwells in conjunction with the automated analysis and classification streamline the analysis process and demonstrate their ability to generalize to other tasks, such as drug detection.