0
0The paper introduce Phoenix, an advanced generative artificial intelligence framework designed to predict single-cell spatial gene expression directly from routine histology images. By scaling training across a massive, highly curated dataset named The Nest, this system successfully overcomes the high costs and slow speeds traditionally associated with spatial transcriptomics technologies. The research demonstrates that Phoenix can generalize exceptionally well zero-shot across diverse, previously unseen biological cohorts, organs, and species without requiring fine-tuning. Furthermore, the framework enables powerful in silico analyses, facilitating population-scale cancer atlases, precise biological hypothesis testing, and the discovery of actionable clinical biomarkers for predicting therapeutic responses.
References:
Tran M, Gindra R H, Putze P, et al. Pan-cancer virtual spatial transcriptomics from routine histology with Phoenix[J]. bioRxiv, 2026: 2026.04. 25.720812.
