Measuring gene activity in individual cells reveals the diversity within tumors but is costly, technically noisy, and often captures too few rare cells. Computer-generated cells could fill these gaps. Here we show, across two independent groups of pancreatic cancer patients, that a deep generative model produces realistic synthetic cells. Explicitly modeling correlations within the model’s compressed description of each cell does not improve preservation of gene-gene relationships over a simpler matched model, which appears to capture this structure implicitly. Nevertheless, the model’s sampling strategy yields cells that faithfully reproduce gene activity patterns, cell-type markers, and cellular organization and compare favorably with established methods. The model trains 100 to 1,000 times faster than a diffusion-based alternative. Synthetic cells blend with real data and support standard analyses, including reconstruction of a precancerous transition. These findings support using synthetic cells to enrich rare populations, benchmark methods, and simulate transitional states in tumors.
The authors express their gratitude to Maite Iglesias Badiola, Cruz Santos Tejedor and Alberto López Rosado for their unwavering support to the groups.
This work is part of the project PID2025-167971OB-I00, funded by MICIU/AEI/10.13039/501100011033. This study has been funded by Instituto de Salud Carlos III (ISCIII) through the project PI22/00492 and cofunded by the European Union (FIS PI22/00492) and Ayudas a la Investigación UFV Grant (UFV2022-23) to V.J.A.L. A.L.R. was funded by Universidad Francisco de Vitoria (UFV) through the Predoctoral Researcher Training Contract (IPF). J.M.S.V. was funded by Universidad Francisco de Vitoria (UFV).
Grupo de Oncología Molecular, Instituto de Investigaciones Biosanitarias, Facultad de Ciencias Experimentales, Universidad Francisco de Vitoria (UFV), Madrid, Spain
Source: Nature - Machine Learning — Read the full article on the original site ↗