️ Episode 167: DeepScence: Detecting Senescent Cells at Single-Cell and Spatial Resolution

In this episode of PaperCast Base by Base, we explore a Cell Genomics study introducing DeepScence, a deep-learning autoencoder that leverages a compact “CoreScence” gene set to identify senescent cells across single-cell and spatial transcriptomics data, outperforming marker- and gene set–based approaches.

Study Highlights:
The authors systematically compared nine published senescence gene sets and distilled a consensus 39‑gene CoreScence panel that is consistently associated with senescence across tissues and conditions. DeepScence models expression counts with a zero‑inflated negative binomial autoencoder whose bottleneck separates senescence‑related signal from unrelated variation and outputs a continuous senescence score that can be optionally binarized. Benchmarking on multiple in vitro and in vivo single‑cell datasets shows that DeepScence achieves higher AUROCs than competing methods, and its scores track experimentally validated enrichment of senescent cells in disease or injury contexts. The method generalizes to spatial platforms including Visium and simulated Xenium panels, retaining strong performance with small targeted panels and across species and tissue types.

Conclusion:
By centering analysis on a robust core signature and a modality‑aware autoencoder, DeepScence provides a scalable, cross‑platform way to map senescent cells and accelerate aging and disease research.

Reference:
Qu Y, Ji B, Dong R, Gu L, Chan C, Xie J, Glass C, Wang X‑F, Nixon AB, Ji Z. Single‑cell and spatial detection of senescent cells using DeepScence. Cell Genomics. 2025;5:10 1035. https://doi.org/10.1016/j.xgen.2025.101035

License:
This episode is based on an open-access article published under the Creative Commons Attribution 4.0 International License (CC BY 4.0) – https://creativecommons.org/licenses/by/4.0/

Support:
If you'd like to support Base by Base, you can make a one-time or monthly donation here: https://basebybase.castos.com/

On PaperCast Base by Base you’ll discover the latest in genomics, functional genomics, structural genomics, and proteomics.