Khurana, T. K.; Wu, L. Y.; Gherardini, P. F.; Moeinzadeh, S.; Mohseni, M.; Yasar, F. G.; Dettloff, R.; Poelma, J.; Sabri, S.; Scaramozza, A.; Li, Y.; Rouzbeh, N.; Elder, N.; Deshmukh, S.; Majd, A.; Gupta, R.; Farahvashi, S.; Thomas, I.; Betts, C.; Charbonier, F.; Roberts, D.; Hsiung, P.-L.; Zargari-Pariset, E.; Yau, R.; Vila, O. F.; Phillips, M.; Xiao, C.; Wang, J.; Zhou, Y.; Adhikari, P.; Taing, M.; Farjami, E.; Javanmardi, B.; Siu, M.; The Cellanome Development Team, ; Valente, C.; Cox, C.; Geiger-Schuller, K.; Turley, S. J.; Rozenblatt-Rosen, O.; Fattahi, F.; Ecker, J. R.; Jones, J. R.; Ga

Dynamic transitions between cell states underlie both normal physiology and disease. However, most single-cell technologies capture only static snapshots. To address this gap, we developed a platform that integrates light-guided hydrogel polymerization with computer vision to generate on-demand compartments around live cells, enabling longitudinal imaging of cellular behavior paired with whole-transcriptome profiling of the same cells at scale. These data link dynamic phenotypes with molecular programs, enabling deeper characterization of cellular states. This approach revealed an adaptive, drug-resistant state in lung cancer cells characterized by potassium channel upregulation and p53-dependent quiescence. In models of adipogenesis and microglial phagocytosis, joint analysis of imaging and transcriptomic data identified key drivers of cellular function that were missed by transcriptomic clustering alone. These results establish the value of paired functional and transcriptomic analysis to resolve molecular drivers of complex cellular behaviors.