Single-cell gene expression profiling of a murine cell line with and without heat stress.

As a proof of concept of our approach to finding differentially expressed genes between cell fates predicted by image-based deep learning, we predicted the fates (survival and death) of heat-stressed cells and acquired those gene expression profiles. We temporally heated the mouse acute myeloid leukemia cell line C1498 at 45°C for 10 min and subsequently performed time-lapse imaging. We picked the imaged cells at -1, 1, 5, 10, and 15 h after heating completion using our developed Automated Live-imaging and cell Picking System (ALPS) and obtained those gene expression profiles. As a control experiment, we heated cells at 37°C instead of 45°C.