Computational modeling of adhesion-based cell sorting in label-free microfluidic platform

Identifying and isolating cells that express desired molecular surface markers is required in a variety of applications in the biological sciences, cell therapy, and medical diagnostics. We develop a biophysical approach for high-throughput and label-free sorting of cells by their affinity for target ligands by molecular surface markers. Our approach consists of a microchannel decorated with periodic skewed ridges and coated with adhesive molecules. To examine the effects of specific and non-specific adhesion on cell trajectories, we conduct small scale experiments in conjunction with computer simulations. We perform both systematic simulations, to find parameters that best align with experimental results, and simulations with new layouts, to explore alternative configurations without the need to conduct a physical experiment. Our specific adhesion model captures the adhesion kinetics and accounts for the molecular binding/unbinding events due to specific interactions of adhesive molecules and corresponding ligands. We find that specific and non-specific adhesion lead to distinct parameters of cells trajectories such as cell interaction time with the ridge, which can be used to identify different types of cell interaction. Furthermore, cell trajectories are sensitive to adhesion level and therefore can be used to sort cells with different ligand expression.