Abstract
Cell population heterogeneity is increasingly a focus of inquiry in biological research. For example, cell migration studies have investigated the heterogeneity of invasiveness and taxis in development, wound healing, and cancer. However, relatively little effort has been devoted to explore when heterogeneity is mechanistically relevant and how to reliably measure it. Statistical methods from the animal movement literature offer the potential to analyse heterogeneity in collections of cell tracking data. A popular measure of heterogeneity, which we use here as an example, is the distribution of delays in directional cross-correlation. Employing a suitably generic, yet minimal, model of collective cell movement in three dimensions, we show how using such measures to quantify heterogeneity in tracking data can result in the inference of heterogeneity where there is none. Our study highlights a potential pitfall in the statistical analysis of cell population heterogeneity, and we argue this can be mitigated by the appropriate choice of null models.
Highlights
groups of identical cells appear heterogeneous due to limited sampling and experimental repeatability
heterogeneity bias increases with attraction/repulsion between cells
movement in confined environments decreases apparent heterogeneity
hypothetical applications in neural crest and in vitro cancer systems
In Brief We use a mathematical model to show how cell populations can appear heterogeneous in their migratory characteristics, even though they are made up of identically-behaving individual cells. This has important consequences for the study of collective cell migration in areas such as embryo development or cancer invasion.
Footnotes
Email addresses: l.schumacher{at}imperial.ac.uk (Linus J. Schumacher), maini{at}maths.ox.ac.uk (Philip K. Maini), baker{at}maths.ox.ac.uk (Ruth E. Baker)
