Mice are not human, why do we focus so much on data in longevity and pharmacology?

(From original article by Andrew Franklyn-Miller)

This really interesting preprint focuses on the molecular basis of aging in mice, particularly the identification of primary aging signatures and their association with mortality rates. The authors conducted DNA methylation (DNAm) analysis in aging mice and identified a primary aging signature with an exponential age dependency, closely reflecting the Gompertz law's description of mortality acceleration. This signature is linked to dynamic instability in the organism's state, driving the aging process in mice.

The study also identified a linear DNAm signature indicative of global demethylation levels, associated with irreversible processes in aging organisms. The linear signature captured the exponential expansion of the state space volume spanned by individual cells within an aging organism, representing the accumulation of irreversible entropy. The authors found that interventions like caloric restriction and parabiosis impacted the dynamic aging signature but did not significantly affect the entropic feature, reinforcing its association with irreversible damage.

Utilizing Principal Component Analysis (PCA)

The research also addressed the traditional approach of developing "aging clocks" or "biomarkers of aging" using supervised learning techniques and the limitations associated with regularization. The study utilized principal component analysis (PCA) to obtain a semi-quantitative understanding of the dynamics of methylation profiles associated with aging and responses to interventions in mice without aggressive regularization.

Correlation and Validation of Aging Signatures

The findings indicated that the leading aging signature, DNAm-PC1, exhibited exponential correlation with chronological age, closely mirroring the Gompertz mortality law exponent. The variance of this exponential feature doubled its rate. On the other hand, the second leading aging signature, DNAm-PC2, demonstrated a linear correlation with age, reflecting global demethylation of mouse DNA. The study validated these findings using single-cell DNAm data and mutual information analysis, reinforcing the link between the linear signature and irreversible entropy accumulation.

Comparison in Mice and Humans and Impact of Interventions

The authors compared the leading aging signatures in mice and humans, identifying qualitative differences but suggesting that both species could be quantitatively described using the same overarching model. The study highlighted the potential of using short-lived species like rodents as models for the discovery and testing of experimental interventions that could significantly affect human lifespan.

Are Mice the answer?

In conclusion, the authors highlight that aging is a multidimensional phenomenon, making it difficult to quantify the effects of life-extending interventions using a singular aging clock. Although regression-based clocks uncontrollably mix the dynamic and entropic features with a bias toward capturing the dynamic (reversible) aspects of aging, particularly in mice.

Mice, with their inherent dynamic instability, can exhibit pronounced reactions to life-extension interventions aimed to reduce the dynamic aging signature, which may only yield small (incremental) effects in humans.