A robust single-cell phenotyping analysis platform to quantify morphological heterogeneity
Embargo until
Date
2018-12-17
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Johns Hopkins University
Abstract
Cell morphology encodes essential information that informs many underlying biological processes and responses to perturbations. Quantification of cell morphology has seen tremendous advances in recent years. However, limitations arise with regards to effectively defining morphological shapes and evaluating the extent of cellular heterogeneity. Using cell and nuclear contours generated using standard segmentation algorithms, we demonstrate the ability to classify cells based on a data-driven approach to determine shape mode distributions. These shape mode distributions provide users with the ability to directly quantify cell shapes and associated heterogeneity without oversimplification, with an effective visualization scheme that relates cell shape morphologies to morphological subtypes.
Description
Keywords
cell morphology, single-cell, eigen shapes, cell phenotypes, heterogeneity