Confocal microscopy imaging provides valuable information about the current expression states within in vitro cell cultures. However, few tools exist to quantify the spatial organization of the cells observed in these images. We present a modular, general-purpose pipeline that extracts cell-specific signal intensities from confocal microscopy images. The pipeline then assigns cell types based on corresponding intensities and quantifies spatial information among cell types through topological data analysis (TDA). We provide an overview of TDA and discuss biological insights which we may gain from applying our pipeline to microscopy images. In particular, we focus on studying the pattern formation of human induced pluripotent stem cell (hiPSC) cultures, which have become powerful, patient-specific test beds for investigating the early stages of embryonic development. By applying our pipeline to images of hiPSC colonies, we are able to detect and quantify changes in pattern formation caused by cell-to-cell signaling and differentiation. We also contextualize our pipeline within a larger effort toward developing quantitative tools for evaluating spatial models.
Note:
The recording is available at https://umass-amherst.zoom.us/rec/share/4NtVcGniXhMPW-pnzEDsHC-UN7N9Cpyk...
https://umass-amherst.zoom.us/j/92966388240?pwd=dXMwUlA0WFFvNUN4ZHdEZndt...
Meeting ID: 929 6638 8240
Passcode: MathBio