Machine Learning Modeling of Metabolomics in Blood Tests For Aging Assessment

#1

Machine learning and data-driven inverse modeling of metabolomics unveil key processes of active aging

https://www.nature.com/articles/s41540-025-00580-4

“Moreover, several studies have reported that endurance exercise but not resistance exercise has a high relevance to aging related questions. In a study by Cao Dinh et al., 2019, it is reported that among 100 old women (aged over 65 years) strength endurance training significantly reduced senescence-prone T cells, which is widely recognized as age-related65, while intensive training showed no significant influence. In another study, Weiner et al., 2019 concluded that endurance but not resistance training has anti-aging effects while examining a total of 124 healthy previously inactive individuals[49](Machine learning and data-driven inverse modeling of metabolomics unveil key processes of active aging | npj Systems Biology and Applications). These studies provide additional support for our body activity index and metabolic network analysis.“

“Furthermore, since natural dementia is a hallmark of aging, we explored its associations with our primary findings—despite physical performance being the central focus of this study. Emerging evidence indicates that in Alzheimer’s disease, brain levels of D-aspartate are dysregulated and neuronal N-acetyl-L-aspartate (NAA) is reduced, reflecting impaired neurotransmission and neuronal integrity71; aspartate serves as an agonist at N-methyl-D-aspartate (NMDA) receptors, and its metabolic imbalance contributes to synaptic dysfunction and cognitive decline72. Peripherally, lower mid-life alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels are associated with increased long-term dementia risk73, while an elevated AST/ALT ratio correlates with poorer cognitive performance and hippocampal atrophy in older adults[74](Machine learning and data-driven inverse modeling of metabolomics unveil key processes of active aging | npj Systems Biology and Applications). Collectively, these findings underscore regular physical activity as a potent, modifiable factor in reducing natural dementia risk75. Together, central aspartate metabolism and peripheral liver enzyme alterations underscore a liver–brain axis in dementia pathogenesis, suggesting novel biomarkers and therapeutic targets.“

1 Like
#2

I freaked out reading the ALT and AST conclusions until I read they are talking about numbers lower than 10 UI/L, the lowest 10th percentile of the population.

1 Like
#3

Yes, also, if you read the paper (it’s a pretty interesting paper!), athletes and people who exercise in general, have a lot of plasticity in their AST and ALT with the levels changing significantly with activity. Keep this in mind next time you have a blood test the day after intense exercise.

1 Like
#4

I am not a big fan of this sort of data-mining study. You take a big enough dataset, with enough variables, and you can find all sorts of correlations. Are they going to be relevant, interpretable or modifiable (i.e. my AST/ALT ratio is high, thus I’m at risk of poor cognitive performance, thus I should work on changing the ratio)? I am doubtful of that.

For the AST/ALT thing, it’s common knowledge (and the authors state it in the paper) that for the majority of people, if those enzymes are high in the blood, it’s because the person has liver, muscle or heart issues.

One thing which I don’t know is whether the levels of those enzymes in the blood have anything to do with your actual NAA/aspartate metabolism in the brain? The enzymes we measure result from cell injury, causing the enzymes to leak out. If you do a heavy workout, they leak out. If you are in a car accident, they leak out. I’ve always considered their presence in the blood as a passive byproduct.

I’m also not a fan of the authors stringing together multiple speculative ideas. For muscle T cells… ok, so older people who resistance train have less senescence prone T cells which is associated with ageing… so does that mean resistance exercise is bad or pro-ageing? It’s pretty obviously not. Same with the other study they cite about telomeres.

This sort of study has the potential to be interesting, using big omics methods and lots of compute power to learn things about health and disease. But unfortunately the way paper publication works is that you have to “find” something and then make lots of claims about it.