September 10, 2025

scAgeClock: a single-cell transcriptome based human aging clock model using gated multi-head attention neural networks

View Full Paper

Ask AI

Bookmark

scAgeClock yields a low mean absolute error of 2 years in age prediction for the best-performing cell type, enhancing accuracy.
Over 16 million single-cell transcriptome profiles were analyzed across 40 human tissues and 400 cell types, providing extensive data for the model.
The observational analysis revealed key aging clock genes involved in mechanisms like translation and programmed cell death.
scAgeClock's findings may facilitate future aging research, with a publicly available model for broader applications.

Discussion

No takes yet. Share an insight, caveat, or question.

This model demonstrates improved biological age prediction in single cells, suggesting implications for aging research and disease risk assessment.

Gangcai Xie (2025) studied this question.

View Full Paper

Ask AI

Bookmark

This model demonstrates improved biological age prediction in single cells, suggesting implications for aging research and disease risk assessment.