Benchmarking single-cell foundation models for aging biology
- 1Artificial Intelligence Research Institute, Shenzhen University of Advanced Technology
Abstract
Single cell foundation models (scFMs) provide representations of cellular states, but their utility across biological questions in aging research remains unclear. We established a benchmark of cellular representations for aging research, evaluating ten general-purpose scFMs, three aging-specific models and conventional methods across five biological questions using more than 2.5 million single cell transcriptomes. Using frozen pretrained representations, Geneformer performed best among scFMs for chronological age prediction and age pseudotime concordance, although 2,000 highly variable genes achieved higher mean performance. Several scFMs captured positive molecular age shifts across three disease contexts, consistent with reported aging-associated changes. SCimilarity performed well for rare cellular state identification across out-of-distribution datasets, exceeding aging specific models and conventional baselines. At the gene level, scGPT showed the highest recovery of reference TF target interactions, including aging-related regulatory hubs. Overall, scFMs supported diverse aging analyses, but performance depended on the biological question, highlighting their utility for rare cellular state identification and regulatory analysis.
Related
- Benchmarking single-cell foundation models reveals an advantage in the low-data regime for cell-type and disease-state prediction
- Harmonised benchmarking of foundation models for single-cell and spatial transcriptomics reveals context-dependent generalisation
- Where single-cell foundation models help: a benchmark of annotation, perturbation and regulatory inference against simple baselines
- scRM: A single-cell foundation model trained on scRNA-seq reads
- Single-cell transcriptomic and epigenomic analysis reveals X-linked sex differences in aging mouse hypothalamus
BibTeX
@misc{ni2026benchmarking,
title = {{Benchmarking single-cell foundation models for aging biology}},
author = {X. Ni and Y. Liang and J. Zhu and L. Zhang and X. Shen and M. Yang and R. Xu and H. Li and R. Ji and C. Jin and S. Ni},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.21.753191},
url = {https://www.biorxiv.org/content/10.64898/2026.09.21.753191v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Benchmarking single-cell foundation models for aging biology
AU - X. Ni
AU - Y. Liang
AU - J. Zhu
AU - L. Zhang
AU - X. Shen
AU - M. Yang
AU - R. Xu
AU - H. Li
AU - R. Ji
AU - C. Jin
AU - S. Ni
PY - 2026
DA - 2026/09/21
PB - bioRxiv
DO - 10.64898/2026.09.21.753191
UR - https://www.biorxiv.org/content/10.64898/2026.09.21.753191v1
AB - Single cell foundation models (scFMs) provide representations of cellular states, but their utility across biological questions in aging research remains unclear. We established a benchmark of cellular representations for aging research, evaluating ten general-purpose scFMs, three aging-specific models and conventional methods across five biological questions using more than 2.5 million single cell transcriptomes. Using frozen pretrained representations, Geneformer performed best among scFMs for chronological age prediction and age pseudotime concordance, although 2,000 highly variable genes achieved higher mean performance. Several scFMs captured positive molecular age shifts across three disease contexts, consistent with reported aging-associated changes. SCimilarity performed well for rare cellular state identification across out-of-distribution datasets, exceeding aging specific models and conventional baselines. At the gene level, scGPT showed the highest recovery of reference TF target interactions, including aging-related regulatory hubs. Overall, scFMs supported diverse aging analyses, but performance depended on the biological question, highlighting their utility for rare cellular state identification and regulatory analysis.
ER -