Making phenomenal insights: CellPhe 2 for enhanced cell phenotyping from time-lapse images

Session

Session 2 - Early Career Researchers Session: the Science of Tomorrow Today

Authors

Laura Wiggins (2, 3), Will Brackenbury (3), Peter O'Toole (3), Alice Pyne (2), Stuart Lacy (3), Graeme Park (3), Beth Cimini (1), Julie Wilson (3)

Affiliations

1. The Broad Institute of MIT and Harvard
2. University of Sheffield
3. University of York

Keywords

Image analysis, cell phenotyping, open-source, software, machine learning, cell biology, time-lapse imaging, fluorescence, ptychography

Abstract text

With phenotypic heterogeneity in whole cell populations widely recognised, the demand for quantitative and temporal analysis approaches to characterise single-cell morphology and dynamics has increased. To address this, our team developed CellPhe, a pattern recognition toolkit for the unbiased characterisation of cellular phenotypes within time-lapse videos. CellPhe has demonstrated its value in quantifying single-cell responses to drug treatment, facilitating imaging of co-cultures and identifying cell-cell signalling events. However, the complexities of biological data sets and evolution of new imaging technologies call for even more advanced analysis solutions. Here we present CellPhe 2, which builds upon our original cell phenotyping toolkit by incorporating new analysis capabilities for population-level modelling, tracklet analysis and interoperability with existing open-source cell segmentation and tracking software. CellPhe 2 is designed to facilitate analysis of complex data sets from multi-channel fluorescence images through to densely populated samples where conventional segmentation approaches underperform.