In this report,we present multiparameter deformability cytometry(m-DC),in which we explore a large set of parameters describing the physical phenotypes of pluripotent cells and their derivatives.m-DC utilizes microflu...In this report,we present multiparameter deformability cytometry(m-DC),in which we explore a large set of parameters describing the physical phenotypes of pluripotent cells and their derivatives.m-DC utilizes microfluidic inertial focusing and hydrodynamic stretching of single cells in conjunction with high-speed video recording to realize high-throughput characterization of over 20 different cell motion and morphology-derived parameters.Parameters extracted from videos include size,deformability,deformation kinetics,and morphology.We train support vector machines that provide evidence that these additional physical measurements improve classification of induced pluripotent stem cells,mesenchymal stem cells,neural stem cells,and their derivatives compared to size and deformability alone.In addition,we utilize visual interactive stochastic neighbor embedding to visually map the high-dimensional physical phenotypic spaces occupied by these stem cells and their progeny and the pathways traversed during differentiation.This report demonstrates the potential of m-DC for improving understanding of physical differences that arise as cells differentiate and identifying cell subpopulations in a label-free manner.Ultimately,such approaches could broaden our understanding of subtle changes in cell phenotypes and their roles in human biology.展开更多
基金We acknowledge financial support from the Packard Foundation and the National Science Foundation grant no.1150588.
文摘In this report,we present multiparameter deformability cytometry(m-DC),in which we explore a large set of parameters describing the physical phenotypes of pluripotent cells and their derivatives.m-DC utilizes microfluidic inertial focusing and hydrodynamic stretching of single cells in conjunction with high-speed video recording to realize high-throughput characterization of over 20 different cell motion and morphology-derived parameters.Parameters extracted from videos include size,deformability,deformation kinetics,and morphology.We train support vector machines that provide evidence that these additional physical measurements improve classification of induced pluripotent stem cells,mesenchymal stem cells,neural stem cells,and their derivatives compared to size and deformability alone.In addition,we utilize visual interactive stochastic neighbor embedding to visually map the high-dimensional physical phenotypic spaces occupied by these stem cells and their progeny and the pathways traversed during differentiation.This report demonstrates the potential of m-DC for improving understanding of physical differences that arise as cells differentiate and identifying cell subpopulations in a label-free manner.Ultimately,such approaches could broaden our understanding of subtle changes in cell phenotypes and their roles in human biology.
