Inferring gene expression profiles of tumor clones.

CLONALGE accurately infers gene expression profiles and spatial distribution of tumor clones, outperforming previous models in prostate cancer analysis.

• Why it matters: Understanding the transcriptional heterogeneity and spatial organization of tumor clones is crucial for advancing cancer diagnosis and treatment strategies, yet tools to jointly analyze genotypes and phenotypes are lacking.
• What they did: The authors developed CLONALGE, a probabilistic graphical model that integrates spatial transcriptomics and DNA sequencing data from Homo sapiens prostate tumors, using Monte Carlo sampling to identify differentially expressed genes among clones.
• The result: CLONALGE reveals clone-specific gene expression patterns and spatial arrangements, enabling detailed functional characterization of tumor subpopulations and improving the accuracy of spatial transcriptomics interpretation.