tage
Elastic Net over 10,487 mouse-Entrez genes, multispecies multi-tissue, scaleddiff variant. Cohort-relative: predict_age runs the tAge cohort preprocessing on the raw RNA-seq counts itself, so a prediction is an age difference against the reference group rather than an absolute age -- zero means "looks like the reference", negative means younger-looking. Name the cohort's species with a 0/1 column among var_names (mouse, rat, macaque or human; absent or all-zero means mouse) and the samples to centre against with a truthy adata.obs["tage_reference_group"] (absent centres on the whole cohort); at least two samples are needed. The published pipeline's SimpleImputer, mean-only StandardScaler and pass-through SelectKBest are folded into the packaged linear layer, and the imputer medians are carried as reference_values so a gene the sample does not measure contributes its training median. Output is months of mouse age: the Elastic Net predicts a fraction of species maximum lifespan and the mouse factor of 48 months is baked into the weights, so a cohort of another species must be rescaled by its own maximum lifespan over 48 (human 122.5 years, rat 50.4 months, macaque 39 years); those constants come from the authors' released tAge package and differ slightly from the paper's Methods, which report 4, 3.8, 39 and 122 years. Released under the MGB Open Access License 1.0: non-commercial academic research use only.
| Predicts | relative age |
| Species | multiple species |
| Tissue | multi-tissue |
| Data type | transcriptomics (relative) |
| Model type | elastic net regression |
| Year | 2026 |
Use with pyaging
import pyaging as pya
pya.pred.predict_age(adata, ["tage"])
Browse every clock in the pyaging Clock Catalogue.
Citation
Tyshkovskiy, Alexander, et al. "Universal transcriptomic hallmarks of mammalian ageing and mortality." Nature 654 (2026): 173-188.
- Downloads last month
- -