This paper introduces scFOCAL, a new computational framework designed to overcome intratumor heterogeneity in glioblastoma by predicting how specific cell populations respond to medication. By integrating single-cell RNA sequencing with large-scale drug response datasets, researchers can identify which tumor cells are sensitive or resistant to various therapies in silico. The study demonstrates the platform's accuracy by predicting that alisertib treatment leads to a shift from neural-progenitor-like states to resistant mesenchymal-like states, a finding confirmed through in vivo xenograft models. Furthermore, the authors utilize scFOCAL to discover a potent new synergistic combination involving an OLIG2 inhibitor and an anti-EGFR drug conjugate. This tool is now publicly available as an R package and web application to help scientists prioritize effective treatments for complex cancers. Thus, the framework offers a scalable method to design personalized combination therapies that target the full spectrum of diverse cells within a single tumor.

References:

• Suter R K, Jermakowicz A M, Veeramachaneni R, et al. Drug and single-cell gene expression integration identifies sensitive and resistant glioblastoma cell populations[J]. Nature Communications, 2026, 17(1): 99.