Dealing with Lyme? From NYT bestselling author and internationally renowned physician Dr. Richard Horowitz comes a revolutionary roadmap to decode complex chronic illness and reverse inflammation from the inside out. This is a deeply hopeful, science-backed program for anyone left exhausted by chronic illness. cangetbetter.com/

Epigenetic insights into extreme longevity in the world’s oldest terrestrial animal, Jonathan 🐢“Notably, we found that lower-entropy regions in Jonathan’s methylome were enriched for the promoters of genes involved in the mitochondrial electron transport chain, and RNA metabolism. This suggests that high-fidelity transcription of the genes in these pathways may be crucial for long-lived species. Our findings support a model for aging wherein the maintenance of low methylation entropy in gene promoters is coupled to efficient mitochondrial energy production, efficient RNA processing, and efficient genomic repair.” science.org/doi/10.1126/scia…

fun fact that's not disclosed in papers: when mice are hunched over/immobile or have large tumors, animal facility staff order them to be sacrificed almost all mouse aging experiments are measuring healthspan/movement span

Here’s your periodic reminder that olive oil consumption actually outperforms most Alzheimer’s drugs! Of course, not all olive oil is created equal. Shop for high-polyphenol, extra virgin olive oil with a visible (recent) harvest date. realsimple.com/best-cooking-…

Sorry, @kfitzgeraldnd, I wasn't able to respond directly to this post because apparently Eric Topol has blocked me from responding to his posts. Funny how some people have no problem dishing out criticism, but aren't willing to approach scientific debate with the same level of integrity they expect of others. Regardless, the major problems here are not complicated. I don't have any huge issues with the study methods per se, it's the interpretation that the results are informative about intervention efficacy. Aside from the well-known error/variance problems with epigenetic clocks, some of which have been pointed out by the same authors, the quality of the underlying studies used here is extremely variable, so it's not an apples-to-apples comparison across interventions. Even more problematic is the circular nature (or more precisely unvalidated construct-validity leap) of the argument. To be clear I'm not saying this is the authors argument, it's the seriously flawed line of reasoning being used to interpret these results as informative on interventions: "we're going to do a study to test whether/which clocks are responsive to interventions that might impact biological age in order to tell us which of those interventions are responsive to the clocks in order to tell us which interventions impact biological age". It's straight up nonsensical. There are a bunch of other issues that the reviewers flagged that further illustrate the problems with the study design that make this reasoning even more problematic. media.springernature.com/ori… Plus, exercise also "underperformed". So, to pick out rapamycin is just dumb.

There has been a lot of discussion around this paper, which I think is a very important one. It exposes something I’ve been concerned about in the epigenetic clock field, and in the aging biomarker field more generally. The fundamental problem is that we take a highly complex, multifactorial, context-dependent process like aging and try to reduce it to a single value. Occasionally we reduce it to a small vector, as with systems-based clocks. Most of these models are linear, and even when nonlinear models are used, we are still imposing a very low-dimensional representation on an enormously high-dimensional biological process. When you train a model on cross-sectional age distributions, it is going to learn broad patterns associated with age in that population. This can be incredibly useful, but it does not mean that the resulting representation captures the full complexity, nuance, or contextual dependence of biological aging, particularly when we are looking at a single tissue like blood. Then we take that representation and ask what happens under various interventions. Much of this data is necessarily OOD. And of course we do this, because isn’t that the whole point of developing biomarkers of aging? When a clock doesn’t respond to an intervention the way we expect, the conclusion is often that epigenetics is the problem. That it is the wrong substrate for measuring aging. No! It’s the canary in the coal mine. The problem is our incessant need to make complex biology human-interpretable. We take one of the most complex processes imaginable and compress it into a number because a number is easier to label, compare, rank, and put in a box. Then we are surprised when that number fails to behave like the biology. Throwing epigenetics out with the bathwater isn’t the answer. The answer is to embrace the complexity. Add more modalities, more inputs, more context, and models capable of learning high-dimensional, nonlinear relationships across them. We need to stop treating interpretability as a prerequisite for validity. A model does not have to decompose aging into a simple pathway or identify a single “target” to tell us something profoundly useful about the underlying biology. If we want to solve aging, we need to build better representations of the biological state and let the models guide us, even when we don’t fully understand what they are seeing.

Statin's adverse effects on muscle may be driven by inflammation—not just cholesterol lowering. This research shows statins can trigger the NLRP3 inflammasome in muscle cells, leading to atrophy, reduced protein synthesis, and cell death via disrupted isoprenoid/prenylation pathways. Blocking NLRP3 or restoring isoprenoids prevented these effects, identifying a potential target to reduce statin-associated myopathy. #Statins #MuscleHealth #Inflammation #AgingResearch science.org/doi/10.1126/scia…

A recent investigation by @addictedtoigno1 uncovered more than 400 published studies that appear to have used the wrong antibody when measuring p16Ink4a, one of the most commonly used biomarkers in cancer and cellular senescence research. The story has generated understandable concern—and, in some corners of the internet, claims that it invalidates the entire senescence field. I don't think that's the right conclusion. In this episode of Longevity Science, I sat down with @NataliaMitin to discuss what actually happened, what the findings mean, and what they don't mean: youtu.be/b2Lqs4lYLRk We discuss: 🔬 How a naming and cataloging error may have propagated through hundreds of papers 🔬 Why the issue raises legitimate concerns about research rigor and reproducibility 🔬 The difference between honest mistakes, poor validation practices, and scientific misconduct 🔬 Why the strongest evidence linking senescent cells to aging comes from independent genetic and experimental approaches that are unaffected by this antibody controversy 🔬 The incentive structures in science that can reward exciting stories over careful validation My view is that this episode highlights two truths that can coexist: 1️⃣ The p16 antibody story reveals real problems in academic biomedical research that deserve scrutiny and correction. 2️⃣ The broader evidence supporting a role for senescent cells in aging biology remains strong and is based on multiple independent lines of evidence, not a single antibody. Science is not self-correcting because scientists are perfect. It is self-correcting because people are willing to question assumptions, revisit results, and follow the data wherever it leads. I believe that current incentive structures in academic career building, funding allocation, and publishing have degraded these historical guardrails. That's exactly why conversations like this matter. #Aging #Longevity #Science #ResearchIntegrity #Senescence #CancerResearch #Healthspan #LongevityScience #P16