I’ve been following prediction markets for the past few years, mostly on geopolitics and economics, and they tend to be more correct (or at least directionally correct, and more so than so-called “experts” in the media), even with a relatively small number of “voters”. I set my critical mass at about 25–50. Below 25, I don’t pay attention. Above 50, I somehow “trust”.
This paper puts the critical mass even lower, at n = 20! “We have shown that increasing the crowd size improves the quality of the data. This effect occurs especially with small n. From about n = 20, the improvements are only very small and rather not worthwhile in relation to the costs since the costs grow linearly with n. For n greater than 20, even a deterioration of the F1 score can be observed, which is mainly explained by the increase of FP.”
The most famous example is, of course, Francis Galton’s Vox Populi: 800 voters were accurate within 1% of the true result: Vox Populi | Nature
I didn’t expect to see DNP on this list. That’s an interesting one. It’s used in the bodybuilder to burn extreme amounts of fat in record time but it can be fatal if you use too much. Also makes you drench in sweat.
I’m not surprised. While DNP is best known as a very powerful fat loss agent, and a dangerous one at the doses used for that purpose, a much lesser known fact is that DNP has been found to extend lifespan in rodents at doses about hundred times lower than used for dieting. Such doses are safe and inexpensive and at such low doses it induces mild mitochondrial uncoupling which may be beneficial for longevity. I’v been waiting for DNP to be tested more for longevity and am glad to see it in the ITP.
and safest in humans and (closest to generic too?) non Cana SGLT2i
(Would help also disentangle whether Cana is unique or whether it could be that the SGLT1 part of Cana is key and not just (or not at all) the SGLT2 part)
It’s probably too early in the research to confidently test rejuvenating small-molecule cocktails, but down the line, I think this will be a very promising avenue for the ITP to test
One of the most effective current rejuvenation cocktails is VC6TF, which consists of valproic acid (HDAC inhibitor), CHIR-99021 (GSK-3 inhibitor), E-616452 (TGFβRI inhibitor), tranylcypromine (irreversible MAO inhibitor that also inhibits the histone demethylase KDM1A), and forskolin (adenylyl cyclase activator currently being tested in the ITP).
Right now we have results on various cocktails showing rejuvenation in human fibroblasts, but we need to know which of these cocktails can also rejuvenate transcriptome and reduce epigenetic age in various mice tissues, without loss of cellular identity. We also need to know the minimum concentrations needed for each component of the cocktail, to make delivery feasible and minimize off-target toxicity.
Follow-up studies are underway to elucidate the cellular machinery that mediates these rejuvenative effects, with an emphasis on the mechanisms by which cells apparently write then later read a “backup copy” of earlier epigenetic information to reset chromatin structures and reestablish youthful gene expression patterns.
^ There is more work being done on this as we speak.
Although I don’t find it a convincing idea that cells contain a backup of epigenetic information, perhaps they meant that as an analogy? I think that cell type-specific epigenetic marks are (obviously) thermodynamically favorable, at least in newly differentiated cells. This is due to cell type-specific complements of transcription factors and other DNA-binding proteins which favor certain loci and create stable methylation patterns at those loci, and this steady-state is obviously favored in the presence of youthful metabolome, youthful proteome, and youthful values of other high-level information like electric and ionic gradients.
Over time, or perhaps even shortly after differentiation, some of those loci-specific DNA-binding proteins are lost, and the youthful patterns of DNA methylation are no longer thermodynamically favorable. This in turn further changes the patterns of DNA-binding proteins, which further changes the DNA methylation. It’s a vicious cycle.
From this perspective, cellular rejuvenation is more like folding a protein than it is burning a CD. If you can push the proteome (and especially the DNA-binding subset of the proteome) and the metabolome towards a newly differentiated/youthful state, you get youthful DNA methylation for free, since those patterns of DNA methylation are favorable under those conditions.
So it’s about finding small molecules which can mimic or induce that initial set of DNA-binding proteins, and perhaps that might be a different set of molecules for each cell type. And as far as essential organs go, you’re only as good as your weakest one, so you would have scheduled pulses of each cocktail for each organ to avoid interactions and minimize extended exposure (in this view you’re resetting the clock and not just slowing the rate it ticks, so pulsing should work just fine).
As I see it the key to methylation patterns is how frequently a gene is transcribed. Hence if you have genes that are rarely transcribed they will be more heavily methylated.
Limits on transcription come from diffentiation, whether the cell’s logic systems require a gene to be transcribed and indicated energy levels (indicated by acetyl-CoA levels).
Hence when energy levels are lower (at a cellular level and often because of mitochondrial damage) then genes will end up more methylated.
This, however, is not a question as to whether there is a “backup of epigenetic information”, but a metabolic question.
That’s also my assumption. The trials are not double blinded I guess so they can see early if one intervention is super successful. Same for CANA and STS in 2020 btw.
Does anyone know when the deadline to submit an application is? And is it really worth applying? I assume that they already have in mind the compounds I’m interested in (taurine, GlyNAC, oral semaglutide, empagliflozin, sotagliflozin, selegiline, fenofibrate, rapa + CANA) so applying might be useless…
The submission deadline occurs once per year on the last weekday in February.
The next deadline is Wednesday, Feb. 28, 2024.
Is it worth applying? Hard to tell, right? Sadly they really don’t share much information on the applications - past or present, unless chosen… It would be nice to know what had been submitted in the past and turned down, every year.
But - they start sharing information, pro and con, on the chosen compounds when they are about half way through the trial…
Thanks. I’ll do the FOIA request. I’ve never done it in the US but non citizens can do it and I’ve have experience of similar requests in the UK, France, and the EU (including up to the European Court of Human Rights HUDOC - European Court of Human Rights ).
That would be fantastic! Let us know how much work it is - and if we need to crowd-source the effort - post a message here in the forums and we can get a group together perhaps, to work on it!
All the proposals (PDF or Word file) submitted to the NIA since January 1st, 2022, as part of the ITP’s “Request for Proposals”,
The list of all interventions under consideration for the ITP’s 2024 cohort,
The list of all interventions under consideration for the ITP’s 2025 cohort.
=> ITP: Which drug(s) will increase median lifespan the most in males by 2030? | Manifold