109 replies to this topic

[OlderThanThou2]

I've been doing 3 runs of 3 days of senolytics in the past 3 months at 6 weeks intervals, while taking 5ml per day of C60OO and 10g mango oil ( providing about 4g of stearic acid ). The logic is to try and renew the cells that have been killed by the senolytics, while keeping the number of stem cells. I try to make sure that the organs have completely recovered between taking senolytics. Don't for sure if this is a good approach. I intend to lower the dosage of both C60 and stearic acid 4 weeks after I've done a 4th run of senolytics, which will be the last.

Edited by OlderThanThou2, 06 December 2020 - 11:31 AM.

[ambivalent]

And rats equal mice?

 

Same researchers, diametrically opposed rodent results.

 

https://pubmed.ncbi....h.gov/21410319/

[ambivalent]

Suppose the above referenced studies on the regenerative effects of fasting on spinal chord injury in rats and mice respectively had been conducted by separate labs. The thought-space generated from observing the other lab's apparently conflicting data would be in contrast with  explanations considered had said data been produced on home soil - or at least the relative weight of the explanations. However, while those scientists will have had some thought to look for loose cables, they most likely would be drawn to: 'what's going in rats that isn't in mice?' Views of the woods and the trees are both legitimate - the within lab and across lab perspectives are both valid and biassed but it is worth trying to hold  both in mind.  

 

The fasting-spinal-chord-injury guys might see Moody's and Moussa's work as further dramatic evidence of differing intra-rodent biological mechanisms. If Kelsey had conducted that Baati/Moussa study himself with those sets of results - he would be of the mind, I'm sure, of either what's different about rats and mice or what did I do differently with between the rats and the mice. But from the interview it seems his thoughts are more of the kind 'nothig to see here, move on' and that despite the unmentioned c60oo mouse leukemia study where his lab demonstrated efficacy with c60oo.

 

Now, what conditions must be satisified in order to dismiss c60's life extension/rejuevenating properties demonstrated in Baati Wistler rat study?

 

Well, rule out poor experimental conditions or small sample size:

 

On the sample size: I found 150 surprising. The only purpose of a sample size this large is to measure accuracy, not effect. if the Baati study translates to mice you're not going to need a lot more half a dozen. And if it fails as it would have, then there is resource to test other conditions or even go back to the rats.     

 

If Baati was experimental error or design, well then we're after that error, whatever that error is. That leaves us seemingly with either fraud or some lab-sabotage - both of which seem more likely than conducting a chance doubling of lifespan in rats (except thats often how science discovers). But that is some accusation - tacitly communicated by those dismissing the study, when no attempt has been made to replicate it. And in fact KMoody's AML study was heralded as something of an endorsement of Baati's, there was a similar albeit contracted effect on life-extension. If c60oo is really a nothing burger, then the albeit weakish mirroring of Baati's study would still be rather flukish

 

There is a far greater endorsement to the Baati study. And it has been here on longecity. Seasoned self-experimenters have demonstrated such outlying effects and even bizarre ones. There were many reports of individuals finding dramtically increased alcohol tolerance (and recovery) in  proximity to taking only a few ml of c60oo. It is tough to put this on a placebo, when the effect isn't commonly demonstrated elsewhere and arrived at seemingly through independent accounts. If this effect is true and real, then c60 is demonstrating extremely high potency in creating a remarkable if not particularly sought after effect. It isn't surprsing that extremely suprising things will happen when ingesting a substance that doubles the life-span of rats. And we had it on here for a couple of years. There are plenty of very unusual, commonly reported testimonials with c60 - longecity has been a round a long time.

 

Moving on from c60oo seems incredulous - with a paper like Moussa&Baati and the subsequent testimonials on longecity, dismissing c60oo with a  99% certainty would appebe ar negligent.   

 

So how to move forward?  

 

Well, without wishing to complacent about about toxicity experiments on rats, given the effects on mice of ingesting light exposed c60oo, further investigation appears mandatory given the large scale consumption of c60oo amidst apparent industry manufacturing complacency. 

 

Proposal: A study testing the toxic effects of light exposed c60oo on Wistar rats. 

 

Outcome 1: Confirmed. This would be a very powerful result. While it seems on the surface almost trivial to demonstrate that which is acutely toxic in mice is acutely toxic in rats. It most certainly wouldn't be to the psychology of the consumer - if that which doubled the lifespan of Wistar rats when properly prepared, poisons them when poorly prepared (or stored) then there is an easier translation (it sure would be to the Wistar community).

 

If c60oo has hitherto been good to you, then you're more likely to align yourself to the rats and in a sense the recent mouse longevity study affirms this mindset:

 

"c60oo didn't offer any life-extending advantage for mice, but was amazing for rats - its been amazing for me so I'm with the rats - I should be less worried about the toxicity of light exposed c60oo because this effect occurred with a species that didn't gain from taking the non-toxic stuff, unlike me and the Wistar rats"

 

If c60oo under certain conditions demonstrates acute toxicity in Wistar rats, then it psyhologically anchors the Baati study - we are more bound to be cautious.

 

Outcome 2: Disproven. Would anyone be that surprised?  The mice die, but the rats are fine. In a thrice we have a result which makes some sense of the contrasting longevity studies, without explaining them. (Needless to say, demonstrating that both rodents species are posioned by light-exposed c60oo doesn't inhibit the claim pristine c60oo reaches parts in rats in doesn't in mice - i.e. that both the longevity results are credible.)

 

Experiment 2

 

Administer pristine c60oo to both Wistar rats and mice. Then at some determined time later, adminsiter the toxic c60oo. Now wouldn't that be interesting?

 

These require pretty short time frames to demonstrate effiicacy: If light exposed c60oo is as toxic in Wistar rats as in mice and pristine c60oo doesn't protect against toxic c60 in rats or mice, then we'll know inside a month.

 

It would also seem sensible given Turnbuckles mito-fusion theory and reference in Baati to having fasting and non-fasting arms, though it is obviously doubling up.

 

Edited by ambivalent, 28 December 2020 - 03:20 PM.

[Jturpin23]

Looking at the article archived here:
https://www.gwern.ne.../2020-grohn.pdf

I find a numerical discrepancy.

 

Eighty milligrams of fullerene C60 (SES

Research, Huston TX, USA) was dissolved in 10 ml of

extra virgin olive oil (EVOO) by stirring for 2 weeks at

ambient temperature in the dark. For this purpose, com-

mercial, high-quality EVOO from Central Italy was

used. The resulting mixture was centrifuged at 5.000g

for 1 h and the supernatant was filtered through a

Millipore filter with 0.25 μm porosity. As the solubility

of C60 varies from 0.6 to 1.2 mg/ml, according to the

olive-oil composition and its moisture content, we con-

firmed that the final solution was at the concentration of

0.8 mg/ml by UV-Visible spectroscopy (absorbance of

toluene diluted solutions at 330 nm).

 

So they are claiming that 80 mg / 10 mL, which is 8 mg/mL, becomes 0.8 mg/mL. In other words they used 10 times as much C60 as they needed and didn't notice the numerical discrepency nor did they note 90% left over C60 in their filtration as is needed to account for the numerical difference. Typically C60 is black and filters are white or clear, so it would be very noticeable.

When I see things like this, I have to wonder how much of this is error and how much is dry labbing. Or maybe the two go together.


 

[Turnbuckle]

Looking at the article archived here:
https://www.gwern.ne.../2020-grohn.pdf

I find a numerical discrepancy.

 

So they are claiming that 80 mg / 10 mL, which is 8 mg/mL, becomes 0.8 mg/mL. In other words they used 10 times as much C60 as they needed and didn't notice the numerical discrepency nor did they note 90% left over C60 in their filtration as is needed to account for the numerical difference. Typically C60 is black and filters are white or clear, so it would be very noticeable.

When I see things like this, I have to wonder how much of this is error and how much is dry labbing. Or maybe the two go together.


 

 

It's clearly a typo, as it should have been 100 ml, not 10. The Baati paper from 2012 claimed they dissolved 50 mg of C60 in 10 ml of olive oil, and after centrifuging and filtering, the solution had a 0.8 mg/ml concentration. People wondered back then about the waste, but at the time the researchers didn't know how much would dissolve. I expect there was copying and pasting involved with this latest paper.

[Jturpin23]

It's clearly a typo, as it should have been 100 ml, not 10. The Baati paper from 2012 claimed they dissolved 50 mg of C60 in 10 ml of olive oil, and after centrifuging and filtering, the solution had a 0.8 mg/ml concentration. People wondered back then about the waste, but at the time the researchers didn't know how much would dissolve. I expect there was copying and pasting involved with this latest paper.

 

I'm not so sure that it is a typo, or if it is, whether it is 8 mg in 10 mL or 80 mg in 100 mL. I did some math.

There were two C60-EVOO groups, one you and one old, with n=8 and n=10 respectively based on Figure 5. So 18 mice given C60-EVOO orally. 

One a day for a week is 7 doses per mouse.  Weekly for a month is, say 4 doses. Fortnighyly for 7 months is say 14 doses. So 25 doses per mouse total.

666 doses / 25 doses/mouse rounded down = 26 mice, so they prepared enough to treat 26 mice.

So 10 mL is plenty, it is a factor of safety of 1.48.

Note also that though EVOO and C60-EVOO was used for oral admininstration, there is no mention of EVOO in the injection experiments. They rather say OO. E.g.,

 

Freshly prepared C60-OO was irradiated with the light of a full daylight spectrum 32 W fluorescent bulb (380–700 nm range) for 8 days

This calls into question whether the problem is light exposure or the method of extraction or some combination of both. Not that I have any plans to inject any derivative of EVOO or OO. I would find it a more compelling warning about quality controls if lab rodents died from oral administration of light degraded C60-EVOO. The conflicts between the papers as it stands leaves me with more questions than answers.

 

Edited by Jturpin23, 08 April 2021 - 01:58 AM.

[Turnbuckle]

 

Note also that though EVOO and C60-EVOO was used for oral admininstration, there is no mention of EVOO in the injection experiments. They rather say OO. E.g.,

This calls into question whether the problem is light exposure or the method of extraction or some combination of both. Not that I have any plans to inject any derivative of EVOO or OO. I would find it a more compelling warning about quality controls if lab rodents died from oral administration of light degraded C60-EVOO. The conflicts between the papers as it stands leaves me with more questions than answers.

 

 

The oil doesn't matter that much, in my opinion. Today you can buy C60 in several oils, including MCT oil, which should be rather more stable than complex oils such as olive oil as it doesn't have any components to react with (though it can still form epoxides). As for EVOO vs OO, the 2012 paper initially described their oil only as "virgin olive oil," and never used the word "virgin" again. They were also aware that while C60 was safe as they prepared it, it was photosensitive. The lead researcher, Fathi Moussa, said this 5 years before--

 

In contrast to chemically--either covalently or noncovalently--modified fullerenes, some C60 derivatives can be highly toxic. Furthermore, under light exposure, C60 is an efficient singlet oxygen sensitizer. Therefore, if pristine C60 is absolutely nontoxic under dark conditions, this is not the case under UV-Visible irradiation and in the presence of O2 where ffullerene solutions can be highly toxic through 1O2 formation. 

https://pubmed.ncbi....h.gov/18217343/

 

 

(Note the typo. I doubt it was in the original French.)

 

C60 is sensitive even to red light that easily penetrates the amber bottles many sell it in.

Edited by Turnbuckle, 08 April 2021 - 11:19 AM.

[Mind]

Another review of C60. Interesting that authors suggest there might be some positive effects of C60, but also raise concerns about low-quality oxidized olive oil. It reminds me about someone else's comment a while back that anyone consuming olive oil should keep it in a box and way from any light because it is so prone to degradation.

 

https://www.longecit...ife-in-rodents/

[trying2survive]

fyi- BuckyLabs finished a mouse C60 study. The C60 olive oil worked very well, but the C60 in mct oil did not. Details:  https://www.buckylab...ongevity-study/

[joesixpack]

Hi, thanks for posting. Your link does not work for me. Here it is again.

 

https://www.buckylab...ongevity-study/

[ambivalent]

All these years later and still c60oo is an enigma. Talk about a game of two halves. How many longevity studies would look anything like this at the midpoint? 

 

Just take a look at the 50% percentile: C60oo is very average. When the group has lost half its numbers and it appears there are 8 or 9 groups doing better, they don't just scrape to victory: in the interval when the mortality rate drops from 50 % to 60% all the other mice are dead. And more to the point the group are convincing winners - that's quite an area under the curve in the 25% percentile between C60oo and the runners up.

 

At the half way stage how confident would we would be that c60 isn't going to do much? Very. How assured would be that this group isn't ging to win? More so. What level of certainty, aftet the 50 percentile data is in, that the group won't produce the last surviving 6 or 7 mice? An Abusrd one. And too, that the last 25 percentile lives 10% longer than the next best group. It is very strange.

 

Well, I am not buying sample size in terms of explaining the second half, there are a lot of mice. It is astronomoically improbably c60 fluked it to nail the top 6. There may have been 80-100 mice when the c60oo group were down to 8. We know how hard the Lotto is.

 

Well, the other explanation takes me back to Agevivo, who has remerged with another longevity mouse study. Two of the three mice lived around 2.75 years over 900 days with the third dying at 22 months and was shown with a necropsy to have a tumour. There was some debate as to how to take the data: a quote from niner:

 

"The problem with statistics is that it has built-in assumptions about normal distributions and randomness. When you understand the data, it's no longer some sort of normally-distributed black box. I think that averaging the tumor case with the two normal cases is inappropriate because these data are not normally distributed- there are two different classes."   

 

So perhaps the same might be said of this study, that there are two subclasses within the sample: one that this amenable to c60oo, another which isn't. Three hundred days is quite an advanced age, so perhaps, several have tumours and the c60oo either accelerates the tumour progression, or simply provides benefits which cannot interfere with the tumours timeline. Those without tumours have a clear run at longevity extension. Although a vastly different sample size, it is curiously similar to Agevivo's study. And the ones that make it here to the end, die quite close together similar to Agevivo's: the first at 22 months, and the last two, from recollection, around 33 months a couple of weeks apart.

 

The other question of course is what about the C60mct oil? Unfortunately no groups on just olive oil or MCT oil. Is this down to the benefits of olive oil, rather than c60? A possibility discussed in the early days of the c60 forum. And in the Baati study, there was benefit from just olive oil.

 

Anyhow, the c60oo group contains a very strange data set, which needs some explanation. Perhaps, there were some unusally unlikely mice within the the first 8 that died, skewing the 50 percentile numbers. The underperformance of the first 50 percentile group is perhaps much more likely a fluke, than the over-performance of the remaining 8 mice.

 

 

 

 

 

 

 

Edited by ambivalent, 04 September 2022 - 11:39 PM.

[ambivalent]

To add some clarification, there seems to be disagreement as to the data/chart interpretation of the buckylab study, on the other stem cell thread. I believe it to be quite unambiguous. Reading the chart at 800 days there is only one group with any surviving mice: c60oo. And the number of those mice alive are 25% of the original group: 4 mice. And when the final non c60oo mouse crashes into the axis, there are 2 or 3 c60oo mice still alive. So at a minimum the top 6 longest living mice are all c60oo, and this from at the half way stage 8 c60oo out of maybe around 100. This is obviously significant. 

 

That the control group has the same average as the c60oo group may have several explanations. 

 

Firstly, it is far more likely that the control group is say a 3SD or 4SD sample, than the 6 c60oo winners are an average group which happened to have fluked their way to the top six (and by some distance). 

 

Or the first 8 of the c60 group were themselves a very freakish sample, skewing what would have been a long lived group.

 

Third, c60oo cannot life extend for some mouse biology state, such as those with tumours. 

 

But it is quite clear there are a number of c60oo rats when all the rest of the 240 mice are no more.

 

 

Edited by ambivalent, 06 September 2022 - 01:58 PM.

[QuestforLife]

I'd agree the results are strange. The C57BL/6 mouse is an inbred strain (so genetically highly similar), but there are various lines of the C57BL/6 mouse that differ widely from one another.

 

That being said this strain is known to suffer from atherosclerosis. 

 

They also engage in barbering behaviour - the dominant mice trim the hair of their subordinates. So it is possible we are seeing a separation in lifespan between dominant and servile mice in the group.  Perhaps this is mediated by C60. I am assuming here that each mice group was housed together.

 

https://en.wikipedia.org/wiki/C57BL/6

 

Edited by QuestforLife, 08 September 2022 - 01:26 PM.

[trying2survive]

BuckyLabs' study link has been updated with a link(at bottom of pg) to a spreadsheet with all the data/graphs/notes etc. 

[Turnbuckle]

BuckyLabs' study link has been updated with a link(at bottom of pg) to a spreadsheet with all the data/graphs/notes etc. 

 

From the data (7 controls and 16 with C60), the C60 group lived 5% longer than the controls -- a bit short of the 90% of the first paper. Group no. 9 did even better than C60, living 7% longer than the controls.

Edited by Turnbuckle, 08 September 2022 - 06:46 PM.

[ambivalent]

From the data (7 controls and 16 with C60), the C60 group lived 5% longer than the controls -- a bit short of the 90% of the first paper. Group no. 9 did even better than C60, living 7% longer than the controls.

 

It looks like 6 controls, which is barely worth doing and more likely to mislead than inform, which is clearly why we should look at the overall data set to draw meaningul conclusions.

 

It is clear the top 7 longest lived mice were c60oo. They had 5 mice over 800 days, 8 over 700 days. There was 132 days between 8th and 9th fatalities. Only 22 other mice from the over 200 mice lived 700 or more days, none lived longer than 800 days. 7 of the 16 c60oo group lived more than 70 days longer than the oldest control.

 

The c60oo 25 percentile lived 9.5% longer than the next best group.   

 

These results are very clearly not best summed by "c60oo lived 5% longer than the controls", that's just not being realistic - I don't understand this resistance, this is an interesting and rather strange data set: that is obvious.

 

For 7 of of a group 16 mice to occupy the top spots out of 250 is remarkable. That is an incredibly confident result - try fluking it - and it is around this observation we should look to anchor our conlcusions, not a 5% longevity advantage of a control group of 6 mice which will obviously be a very low confidence result.

 

It is far easier to achieve a few untimely, early mortalities and skew data than it is to fill the top 7 out of a 250 sample. Which is not say the first half are skewed, there might be a legitimate divide in the data.

[Turnbuckle]

It looks like 6 controls, which is barely worth doing and more likely to mislead than inform, which is clearly why we should look at the overall data set to draw meaningul conclusions.

 

 

 

There were 7 controls. It was originally 8, but one was "let go." BTW, are you associated with a company making or selling C60? I ask as you seem to be doing contortions to make the data say something it doesn't say.

Edited by Turnbuckle, 08 September 2022 - 09:07 PM.

[ambivalent]

There were 7 controls. It was originally 8, but one was "let go." BTW, are you associated with a company making or selling C60? I ask as you seem to be doing contortions to make the data say something it doesn't say.

 

I counted only 6 data points for the control group on the 'individual graphs' tab.

[Turnbuckle]

I counted only 6 data points for the control group on the 'individual graphs' tab.

 

You need to look at the actual data. https://docs.google....RmJWPeA/pubhtml

[QuestforLife]

The groups are small enough individually that there is a large random element, hence it would take a powerful intervention to show up in the lifespan data. Especially as most of their groups have multiple interventions. Comparing small group interventions against small group control is going to be pot luck.

But the fact the last 7 of 245 survivors was C60oo is insanely unlikely.

Assuming equal chance of death in all individuals across all groups, the chance of this order is 16/245 * 15/244 *...*10/239 = 0.0000001186459247 %

Anyone want to ask Josh Mittledorf for his take; he's a statistician?

[ambivalent]

I will check.

 

But let me clear you really should retract those groundless remarks. To suggest I am some c60 seller because I insist on pointing out a glaring observation in the data, is rather like accusing the guy of insisting the Emperor is naked of owning a fashion boutique. 

 

Honestly, I find your stubborness staggering - I didn't make direct reference in my post to our previous interactions out of good grace and respect. However, since the insults are flowing, let us begin:

 

AB: "There is no other way to interpret the chart, that when all the other mice are dead around 40% of the c60oo group are alive. It is unambiguous. It is not my statement, it is a statement of the chart representing the study data. If it is BS, then of course all of this is for nothing - I am assuming it isn't." 

 

TB: The chart makes no such statement. The way to compare them is by area under the curve, and by that method, the control seems slightly better. There is no point in continuing to beat this dead horse. It is just cluttering up the thread. 

 

You were wrong, you knew you were wrong. Rather than admit it, or simply say nothing, you default to ad hominem. And it is so trivial. The data proves my statement, and nullifies yours. I don't take credit, it is a basic chart a 16 year old would understand, which is why I don't believe you didn't know I was right all along. 

 

To suggest that I am the one with an agenda, trying to contort the data, when you quite clearly refuse to see the most basic interpretation while obviously a man of science, who doubtless looks at dozens of charts each day - it is not my integrity that should be questioned here, but yours. 

 

The only reason why no one else had spoken up and reaffirmed what was so obviously true, is because they don't want to piss you off because you can be very grumpy and might ignore their protocol questions, as you did mine the other day.

 

You should have been all over these results, wondering what it implies. Honestly, I don't get it. It is obvious that comparing the c60 data to weak control data and suggesting we move on and ignore the fact that the 7 longest lived mice - from group a of 16 - out of 250, which is so absurdly improbable especially, after the first 8 were lost early and given the margin of difference of the long lived, is utterly non-sensical. Not to view this as the relevant data but instead to repeat the comparative data line with the controls when 1. the control data is a a low confidence sample 2. the c60 is divided into to very distinct extreme data sets. 

 

You might also wish to remember that I linked the research finding that fusion occurs during fasting, which you then took to explain the Baati study, counter productive wouldn't you say, if I were c60 vendor? 

 

My motive is that I am curious, here, you are not and the reason you are not is you don't wish to discover anything that might contradict with your theories.

 

You state your theories as fact, when they are not - no one calls you out on this, but I am sure they see it. The personal results I have always taken as fact as much as one can because I have always trusted your honesty and scientific integrity.. Your ego, though, has seems to have become enlarrged, I struggle to explain it any other way, not held in check by other bright guys who once inhabited this place - it is such a trivial thing to become so stubbon over, I don't even believe that this undermines your theorem and in fact an interpretation might support it. I am simply as always trying to understand, and learn, that as always been attitude here - and I certainly never punch above my weight, your knoweldge is vastly superior to mine, but where I can hold the feet of ideas to the fire I should and would. And you should absolutely encourage anyone to challenge your assumptions, that's how we learn, that's good science, but you don't. We're here for the same purpose, we should always operate under the assumption that any biological theorem is wrong and so keep an open mind.

 

Be courteous to those who courteously contrradict you and show magnanimity when they're right - it seems that you have reached a point where you can never be wrong, so much so, that when you make a rather throwaway remark which happens to be incorrect, you'll defend it to the last even when it flies in the face of sense.

 

To have have acheived so much, to be rightly held in such esteem, yet to be so trivial over such triviality, is utterly baffling: you should be smiling while smoking a cigar. 

 

 

 

 

 

Edited by ambivalent, 08 September 2022 - 11:20 PM.

[ambivalent]

The groups are small enough individually that there is a large random element, hence it would take a powerful intervention to show up in the lifespan data. Especially as most of their groups have multiple interventions. Comparing small group interventions against small group control is going to be pot luck.

But the fact the last 7 of 245 survivors was C60oo is insanely unlikely.

Assuming equal chance of death in all individuals across all groups, the chance of this order is 16/245 * 15/244 *...*10/239 = 0.0000001186459247 %

Anyone want to ask Josh Mittledorf for his take; he's a statistician?

 

I dont know how many mice were left when the C60oo 16 were cut down to 8 but might have been around 100. OK say we have 8 lotto numbers out of 100 balls and all seven get the prize. 

 

100/8 * 99/7 * 98/6 * 97/5 * 96/4 * 95/3 * 94/2  

 

Which looks about 2 billion to one This is not the only statistically meaningful result. 7 out of the top 8 would be aroud 250 million - 1. 

 

But we're not just talking ranking obviously, distance matters too, and they were not winning by a whisker, which adds to the confidence and ramps up the 2 billion figure. If we assumed they were all random variables with the same distribution then to get the seven with distance over the rest is obviously even more unlikely. 

 

This has been such a stupid and unnecessary spat. The comparison with controls here is next to meaningless; the above data clearly isn't.

Edited by ambivalent, 08 September 2022 - 11:52 PM.

[Turnbuckle]

 To suggest I am some c60 seller because I insist on pointing out a glaring observation in the data, is rather like accusing the guy of insisting the Emperor is naked of owning a fashion boutique. 

 

 

 

 

I merely asked the question, and it doesn't seem that you've answered it. It's not unknown to have people here arguing the value of some supplement and it turns out they have some connection to the retailer. As for the observation you made, it is not glaring. There is no reason whatsoever to believe that it is not random noise except that you wanted C60 to do better.

Edited by Turnbuckle, 08 September 2022 - 11:21 PM.

[ambivalent]

I merely asked the question, and it doesn't seem that you've answered it. It's not unknown to have people here arguing the value of some supplement and it turns out they have some connection to the retailer. As for the observation you made, it is not glaring. There is no reason whatsoever to believe that it is not random noise except that you wanted C60 to do better.

 

I have been here a long time, so I have some form. It was less of a question, more of a slur.  And secondly, I've not been some big c60 advocate, I don't wax lyrical over it. Other stuff, more so.

 

And besides, the primary challenge I made was on stem cell depletion, when it had been clear that people, some old, had taken it a very long time and not fallen off a Dorian Gray cliff. This is scientific curiosity, and above anything else a reasonable challenge. That should be the basis of your assumption, unless my position becomes unreasonable, which it didn't nor hasn't. 

 

Turnbuckle, why would you possible believe me highlighting - and I am bored of repeating this - that the seven longest lived mice out of 250 were all on c60oo must be agenda driven. I mean, the motive of the person trying to not to pay attention to this incredible statistical outlier, should be called into question - not the one pointing to it.

 

I don't argue against your work, I have and do, when I do, try to take c60 with stearic acid, I think your are unquestioningly onto something. But I don't feel certain your theories about c60oo are consistent or complete as I've mentioned: that's it. 

 

I am curious as to whats going on here, why the MCT fail? Maybe, c60oo for mice is just about the olive oil. Why too the dramatically different sub data sets? Are mice on c60oo different to rats, if so why? This is interesting, why not be curious? 

Edited by ambivalent, 08 September 2022 - 11:55 PM.

[Turnbuckle]

Yes or no, ambivalent, do you have a connection with a C60 retailer or manufacturer?

[QuestforLife]

I dont know how many mice were left when the C60oo 16 were cut down to 8 but might have been around 100. OK say we have 8 lotto numbers out of 100 balls and all seven get the prize. 

 

100/8 * 99/7 * 98/6 * 97/5 * 96/4 * 95/3 * 94/2  

 

Which looks about 2 billion to one This is not the only statistically meaningful result. 7 out of the top 8 would be aroud 250 million - 1. 

 

But we're not just talking ranking obviously, distance matters too, and they were not winning by a whisker, which adds to the confidence and ramps up the 2 billion figure. If we assumed they were all random variables with the same distribution then to get the seven with distance over the rest is obviously even more unlikely. 

 

This has been such a stupid and unnecessary spat. The comparison with controls here is next to meaningless; the above data clearly isn't.

 

Lies, damn lies and statistics, eh?

 

My calculation was just the chance of that final order in the last 7 places, BEFORE the lifespan study had begun. It is correct within that narrow definition, the question is whether that is relevant here. I might have a further look at the data if I get time later.  

 

I've also emailed the data over to Josh Mittledorf, to see what he thinks. 

 

I think the use of other intervention groups as controls is perfectly valid. But I'd like to see some numbers on that from someone who knows what they're doing. 

[QuestforLife]

I did a few calculations. Caveat, it has been a while since I looked at statistics.

 

The increase in mean life expectancy from C60oo feeding didn't look significant statistically. But I had trouble calculating it because there are some very short lived animals that skewed the mean of the whole group.

 

But I am happy with the calculation for the number of mice surviving into the last 25% (above 3rd quartile).

 

This was 0.2525 of the total non-C60oo fed group (near enough the 25% we would expect)

And 0.5 of the C60oo group.

 

Putting this into a Z statistic calculation (https://www.ztable.net/) for 16 animals of the C60oo group gave a Z statistic of 2.81, which on the table is 0.99752.

 

This means there is only a 100*(1-0.99752)=0.248% chance of this result happening by chance (in a normal distribution).

 

Using the Z statistic is appropriate when you are looking at a small result that differs from a large population result, so I think it is correct to use the whole non C60oo fed group as that population. But others with more statistical skills than me might disagree. 

Edited by QuestforLife, 09 September 2022 - 11:43 AM.

[Turnbuckle]

 

Using the Z statistic is appropriate when you are looking at a small result that differs from a large population result, so I think it is correct to use the whole non C60oo fed group as that population. But others with more statistical skills than me might disagree. 

 

 

Yes. Anyone with statistical skills would disagree. It reminds me of an early ad hoc C60 experiment by a member here. The results seemed terrible even though there were no controls, so several members wanted to use the first to die as a control. They convinced themselves by arcane reasoning that it was legit.

Edited by Turnbuckle, 09 September 2022 - 12:20 PM.

[ambivalent]

I am sure your statistics are less rusty than mine, Quest.  I am not too sure what you have done, here. Your Z is the population minus c60oo and your sample the c60oo group? How is your 4 the quartile feeding in the calculation?

 

A calculaton that would also be interetsting is how likely our sample is not some normalised distribution and how that might be calculated: if say there were 100 mice in the group with two subgroups equally separated.

[QuestforLife]

I am sure your statistics are less rusty than mine, Quest.  I am not too sure what you have done, here. Your Z is the population minus c60oo and your sample the c60oo group? How is your 4 the quartile feeding in the calculation?

 

A calculaton that would also be interetsting is how likely our sample is not some normalised distribution and how that might be calculated: if say there were 100 mice in the group with two subgroups equally separated.

 

The way you do the Z statistic for the 3rd percentile is to compare the proportion of mice in non-C60oo group (202 animals) that live beyond 3rd quartile of lifespan in this population (664.75 days) i.e.51/202 animals, to the proportion of the animals in the C60oo group that live beyond this same point, i.e. 8/16 animals. 

 

Then you feed in these proportions and n=16 into the Z statistic calculation and you get the answer I showed. 

 

This purely judges the likelihood of living into the final 25% by chance (normal distribution) vs. what happened in the C60oo group. For 8 of 16 animals to do that in the C6000 group is only 0.248%. 

 

When I tried to do the same for the 2nd percentile (50%) it did not seem to be statistically significant. Which is what you would expect simply from looking at the graph: only 9 of the 16 C60oo mice lived past half way, but 8 of those made it into the last quarter. 

 

The main problem I have with this analysis is that all the mice seemed short lived. The longest lived mouse (a C60oo one) only lived 838 days, which I think is about average for this strain. 

Edited by QuestforLife, 09 September 2022 - 01:31 PM.