Science: Health: Longevity
multi cohort stepped longevity research   (+1, -2)  [vote for, against]
divide a few hundred mice into age groups, then calculate p values from each group, showing trend as well as different-age efficacy

I thouight of a way to velocitize longevity research. Many drugs and chemicals have been measured at a few studies as causing greater longevity. Follow up stuidies, as well as new studies could use this protocol

Creating a group of 40 cohorts (groups) of 11 mice each, where each cohort is a different month of age quantifies longevization effects of new molecules more rapidly

thus there are 11 mice at 39 months of age 11 mice at 38 month of age 11 mice at 37 months of age continuing to 11 mice at one month of age at the beginning

Giving each of these cohorts a longevity molecule then noting the p value of any longevity effect at a group of 11 causes rapid data gathering as to effectiveness

I may have read that 8 mammals are sufficient to generate a p value at .05 thus 11 mice have higher validity

Trending such as when month 28 to month 37 all show a trend to greater longevity improves the p Value further, while bringing the welcome extra data that the molecule might work on the already mature. Different, yet valid metaanlyses are available from each entire multi cohort is also producible

One of the advantages is that you might be able to test new longevity molecules in months rather than years, because of the small cohorts of elderly mice.
-- beanangel, May 31 2017

Note that this differs from something similar to "we got a lot of middle aged mice, then gave them a molecule, made a graph, and determined a p value."

One of the advantages is that you might be able to test new longevity molecules in months rather than years, because of the small cohorts of elderly mice.

Also it is a commercial product! There are places that maintain mouse breeds as well as genetics. These could improve the quantification math described here then create Ready to research mouse multipaks (multiage distributions) as an list item
-- beanangel, May 31 2017


Yes, this is brilliant, because obviously scientists in present don't use any sort of statistical analysis. If only they had your insight. [+].
-- MaxwellBuchanan, May 31 2017


D'oh - [-]!
-- MaxwellBuchanan, May 31 2017


I do not know [mb] wouldn't it be great to screen say, all 1450 FDA drugs on last six months of lifespan mice to find p <.05 at that group. Just a few months to find a lot of immediately prescribable longevity drugs.
-- beanangel, May 31 2017


No, it wouldn't. At p<0.05 you would just find several hundred flukes.

Researchers who want to use vertebrates in any kind of research already have to do a rigorous assessment of the statistical power available from a given number of animals. So this is either WIBNI, let's all, or WKTEBPWAKWTTA.
-- MaxwellBuchanan, May 31 2017


" I may have read that 8 mammals are sufficient to generate a p value "

All mammals can generate pee.
-- normzone, May 31 2017


the problem with this idea starts with the fact that all the mice have already been cured
-- theircompetitor, May 31 2017


//40 cohorts //

So, four legions of mice. Two mouse proconsuls in command. It has a certain something.
-- pertinax, Jun 01 2017


Well, 1/20th of 1450 is just 73 false leads. at p<.01 it is just 15 false leads. follow up studies that go amiss on just 15 molecules is not so terrible. Think of 5% of the earths population (365 million) on longevity drugs that cause 30% greater longevity. That represents about 100 million additional human lifetimes of existence. That is about twice as many lives saved than all the casualties of WWII (or three Stalins!) I strongly support this research.

Currently deprenyl has been measured at three mammal species at about 20% greater longevity, N acetyl cysteine is near 30%. Metformin varies between 5% and 37%. Highly unusual rapamycin is 60% (!)

Once the FDA pharmacopoeia was analyzed for longevity effects, new drugs for other conditions might be formed around longevity effects so they could compete with the generics.
-- beanangel, Jun 01 2017


//N acetyl cystein is near 30%// I'll bet cash money that NAC does jack diddly bugger all in human longevity. I can take the cash now if you like because there's trials, more trials and meta reviews of trials showing antioxdants (that word needs to go away) don't do anything. Mucus and paracetamol overdose are the only uses of any antioxidant anywhere in medicine. Grr. If you have an oxidation problem, I'm happy to sell Oxygen depleted air.. anyhow, why mop up oxidants with a disposable molecule like NAC when you can reduce the source with a little DNP? It would make a fine treat for the elderly and stop all that complaining about the winter fuel allowance.

Anyhow, we have a lot of people on all the approved drugs, they're all out there doing the experiment in humans right now. I'm sure a couple of compounds will start to shake out in the stats. In the mean time, there's no need to start getting a massive animal bill. Mice are sucky metabolism and ageing models. So are worms and all that. If you want to get old mice faster, well, there's ageing model mice, with mutations to speed that up.

Here's a thought, why don't we do it the other way around? Look for compounds that speed up ageing. Success shortens the experiment! Then work out how it works and do the opposite! Simples.
-- bs0u0155, Jun 01 2017


// reduce the source with a little DNP?// Yes, but the question is, how little? In [sucky] mouse models, the DNP dose for extended lifespan was quite low, if I recall - much lower than is used for weight loss.
-- MaxwellBuchanan, Jun 01 2017


they went with 1mg/litre. That level will still bioaccumulate a little. Maybe 10uM? More than enough for shaving the peaks off any membrane potential spikes, maybe tipping slightly sick mitochondria over the edge to degradation? A bigger dose would still do that, but would also mimic caloric restriction to an extent.
-- bs0u0155, Jun 01 2017


Ah yes - if I remember (can't be arsed to re-find the paper), that was in drinking water? Anyone know how much a mouse drinks?
-- MaxwellBuchanan, Jun 01 2017


Divide a few hundred RESEARCHERS into age groups, then calculate p values from each group, showing trend as well as different-age efficacy ?

Throw a party for the last ten surviving mice er ahh researchers.
-- popbottle, Jun 01 2017



random, halfbakery