Replicating Baati Through Community Science

I noticed this posting and want to comment on it:

This forum is for participants in the "Mprize@home C60" experiment. Longecity is supporting an open science initiative with the goal of replicating the C60 in olive oil experiment conducted by Baati, et al. Experimenters will log their actions and post pictures of the ongoing experiment.

I apologize for always being the one to suggest that an alternative way of doing things might be preferable, but here I go again.

The idea of having an open forum where participants post threads discussing their mice and pictures of them seems good at first glance, but the problem is that it could encourage something of a competitive atmosphere, where different owners are trying to get their mice to be healthier or live longer.

This is why I believe it is essential to not have any mouse owner know about how the other owner's mice are responding. We can't prevent some chatter, but I believe it should be highly discouraged so that one owner's experiences don't influence any other owner's actions.

In my original suggested protocol, which can be found her: http://www.longecity...687#entry518687
I suggested using an online survey tool. The advantages of such a tool are many:

• Standardized questions can be presented to each participant
• Reminder emails can be sent to each participant for each stage of the study
• The data is automatically tabulated in a standard form
• No participant sees anyone else's responses
• No one who is not a participant can add any spurious data (in this case, unsolicited postings)

He he I was surprised not to see it in this very double-blind oriented thread. Thank you for at least bringing the point to the discussion. My belief is that the interest of Mprize at home is not merely an experiment, and that here a tradeoff must be done with other goals: creating a community atmosphere allows to:
- have anyone logged-in view the ongoing experiments -- be like the researcher doing a lifespan test, discuss about science, healthy life extension and other topics.
- test the concept of "open science" where openness and interactions improve the quality of experiments: In my experiment I was reminded to double the dose/kg from rats to mice; here, will such ongoing remarks from anyone help make a scientifically-speaking interesting experiment?
- show that healthy life extension science is concrete (whether or not in the end treated animals live longer or not than the controls) and close to us (not in a parallel dimension that we should ignore in our 'normal' life). Help develop a critical point of view (especially in the animals do not live longer than the cotrols). Explain with a visible, concrete and lively exemple that this is a needed step for applied biogerontology
- educate with the preparation of experiments (we are clearly in ;-), the differences between anecdotes, statistics, and careful observations
- perhaps create some vocations. At least create something inspiring for all.

Edited by AgeVivo, 21 August 2012 - 06:42 PM.

I have 2 questions:

Question1: now that we have started to design the roles (persons ABCD), may I be a participant? Is it compatible with a double-blind approach?

Pros and cons I see:
a)- With the process that has been described I won't know what is placebo and what is treatment so I would be interested
b)- I have 3 mice at home already (I have had many mice in the past) and I feed them c60. I don't think I could mix the samples.
c)- I have some c60 already for them -- not from the same source so I don't know if it should be the same color or not; but there is a risk by manipulating the treatment/placebo that I guess what it is, without desiring to guess.
d)- It makes a little more N
e)- It could help me "live the experiment" like other pet owners to better advise them
Due to point c, I imagine that the question should be asked to the research who makes the placebo, when he receives the C60? and to the other person of trust who will randomize perhaps.

Question 2: Should pet owners be allowed to buy C60? (or oilive oil; just kidding )

Sincerely
AgeVivo

Question1: now that we have started to design the roles (persons ABCD), may I be a participant? Is it compatible with a double-blind approach?

Question 2: Should pet owners be allowed to buy C60? (or oilive oil; just kidding )

I don't see a problem with you being a participant. You should keep giving your mice C60, but get some new mice for the Mprize experiment.

I don't see how we could stop pet owners from buying C60, if they're interested in it. They may want to use it themselves, or give it to some other animals. The nature of the experiment should be made sufficiently clear to them that they know that if they deviate from the experimental protocol, they will wreck the results for everyone. If they decide that they just have to give C60 to a placebo animal, there should be an 'escape clause' in the protocol so that they can drop out, or even remove one animal out of several without messing up the results. It would be far better for us to lose a few animals than to have someone give us bad data because they were afraid we'd be mad at them.

I don't see how we could stop pet owners from buying C60, if they're interested in it. They may want to use it themselves, or give it to some other animals. The nature of the experiment should be made sufficiently clear to them that they know that if they deviate from the experimental protocol, they will wreck the results for everyone. If they decide that they just have to give C60 to a placebo animal, there should be an 'escape clause' in the protocol so that they can drop out, or even remove one animal out of several without messing up the results. It would be far better for us to lose a few animals than to have someone give us bad data because they were afraid we'd be mad at them.

+1
Agevivo's mice are already showing signs of being younger than they normally would be and pet owners love their pets to the point where they will give controlls C60oo.
Its important that they dont feel they have to hide this fact from us and that this data is valuable as it gives us more data on the effects of C60oo started at different ages.

Agevivo's mice are already showing signs of being younger than they normally would be

Thank you Logic but woops. The signals are nice but weak so far and we know nothing about the trend it is taking. *If* they were to die in the next month I would think that ok c60 is perhaps not toxic but if tt didn't obviously extend lifespan. So... let's see.

SmithX, one other thing I want to make clear, information to which you might not have been privy to thus far is that our supplier of C60 is providing us with up to 100 free samples of C60oo. These are packaged in small bottles containing 30ml, enough to dose ONE rat or mouse in a similar manner to the Baati study. Having all of the samples in one size bottle simplifies things to a degree for the supplier and others involved in running the experiment.

So, if one of our experimenters has 3 rats, then they will receive 3 bottles of the material (C60 or Placebo). One for each rat. If they have 5 rats, they will get 5 bottles. Therefore, we wouldn't be able to randomize all the single bottles and put them in a box and send them to AgeVivo (to mail to the experimenters). If that was the process then some experimenters with multiple rats might receive bottles of both C60 and Placebo, no one would know which is which, they might get mixed up, what-not.

_______________________


I think we can trust AgeVivo to participate in the experiment.
_______________________

I think having open reporting in the forums is one of the trade-offs with doing a community science experiment. There is a risk of the experimenters sharing information or influencing each other in a way that might negatively affect the results. I think the way we should handle it is to continually emphasize protocols in the forum with the participants.

I have created my logbook for Mprize at home. We are only 3 so far to have created a logbook, but others told me they plan to do it this week end and it is intimidating for newcomers to directly log in and create a thread.

I have made a simple description of "What is Mprize at home" for someone totally new here: http://www.longecity...567#entry530567 I wrote 2-3 years in terms of time by which we will undisclose results; if we start with 9 months old animals and start treatments at age 10 months // but I could rectify the post.
I hope you will like it

Edited by AgeVivo, 22 August 2012 - 07:38 PM.

we wouldn't be able to randomize all the single bottles and put them in a box and send them to AgeVivo (to mail to the experimenters). If that was the process then some experimenters with multiple rats might receive bottles of both C60 and Placebo, no one would know which is which, they might get mixed up, what-not.

Well detected.
- One option is that I prepare groups of pet owners for which an identical number of bottles is to be sent. For example, if persons F,G,H,I should receive either placebo or treatment for 4 pets each, I will receive four bottles numbered 5 from the researcher, four numbered 6, four number 7, four number 8. I won't know which numbers correspond to treatment or placebo.
- Another option is that I find someone else (...) to keep the bottles at home with a label of what it is, and send them to pet owners after he has removed the label.
Any preference? Something different?


About colorants: the researcher working on colorant told me that he has found something that may work. Should someone exchange with him? Should I propose him to meet the researchers of the paper? There is no urgency: I guess that when he will have the olive oil and the C60oo he should be able to know how good it is and try to improve it if needed.

Hm. that does complicate things a bit.

Whomever prepares the samples could prepare a certain number of ziplock bags with 2 bottles, some with 3, etc. and pre-label them, but you might run out of bags with the correct number of bottles too soon, depeding on how it works out. That would be the best way to do it though.

I have received two packages by mail

• one was sent by Mind and contains the olive oil that carbon60 told us he uses
• the other one was sent by carbon60 and contains
  
• 10 bottles with some dark liquid inside -- in principle the c60 in olive oil produced by carbon60. I haven't opened it more than on the picture so I won't know whether people receive placebo or treatment
• 10 empty bottles
• 20 syringes of 1 ml, to dose the right amount on solution to put on bread, for rats

Here are the corresponding pictures:

I have taken the 10 empty bottles out (they were packaged together) and opened them:

Then I have washed them well with water, rinced them well, washed them well with water and soap, rinced them well. Then closed them and packed them together in plastic again, and have put everything in a box. I will now bring the box to the retired researcher who will create the placebos.

Important Note: In the proces I have put some water on the nice labels that carbon60 had put on the bottles and the labels have started to blurr (sorry carbon60). So
- I have even not looked at them ('seen' but not looked'), and I have not taken photos, so that I won't know recognised them
- I will tell the retired researcher to also wet the labels of C60-containing bottles (sorry carbon60)
=> so that no participant can know whether he received a treatment of placebo.

Ok; I did a lot of work getting the word out to rat and mice owners.
I was going to buy cages and rats or mice and take part in this trial.
It would seem that suggesting that one person could sucessfully keep control and C60 rats without knowing which are which and that suggesting small daily doses rather than large weekly doses will get you ignored like a fucking leper aroud here!?????
WTF!?!???

If you want to ignore me thats fine, but ignorig Maureen is just plain bad manners!!!!!!!!!!!!!!!!!!!
http://www.longecity...my-pet-ratties/
In fact I am sorry that I took the trouble to bring the above thread to the attention of the relivant people as my doing so seems to have got it ignored too!?

Hi Logic,
I am not sure I understand your post. Perhaps it is directed at me, I don't know
- I am very busy currently so everything like answering to PMs or to Maureen takes me one or two weeks, rather than one or two days. If you have time, please don't hesitate to take the role to accompany Maureen (and thank you in advance)
- One thing that has slowed down things a little is that, as discussed internally with Mind, we won't have enough money to pay for the equipments
- Another is that for now we have figured how to buy aged animals in the US only
- Another is that no pet owner has yet opened his logbook on LongeCity yet; if I had had more time I would have accompanied them more and it would have accelerated things a little, but I guess everyone is a quite busy
In fact The project is doing well and Those things are natural parts of the project after all . Lack of money, lack of time (I have a more-than-full-time job), nothing exceptional here. My mice are looking good, that is the best news so far.

What about my objection to the public logbooks and the concern that this could influence results by allowing participants to see what others are experiencing. I am concerned that either of two things may happen:

- Participants will want their results to be similar to ones which other people posted
- Participants will want their results to be better than ones other people posted

Either case could cause severe problems in the data.

Another question: could you specify how the samples are being handled in terms of randomization and delivery?

Also: I believe that all the labels should be taken off or covered with other labels which just have a number on them (1, 2, 3 etc.) corresponding to a spreadsheet decoding which are which.

What about my objection to the public logbooks and the concern that this could influence results by allowing participants to see what others are experiencing. I am concerned that either of two things may happen:

- Participants will want their results to be similar to ones which other people posted
- Participants will want their results to be better than ones other people posted

Either case could cause severe problems in the data.

I am in complete agreement with this. The public logbooks will essentially break the blinding of the experiment. We seriously need to re-think them.

Also: I believe that all the labels should be taken off or covered with other labels which just have a number on them (1, 2, 3 etc.) corresponding to a spreadsheet decoding which are which.

Good idea. The blurred labels may give the impression that a lack of care was taken in sample preparation.

Also: I believe that all the labels should be taken off or covered with other labels which just have a number on them (1, 2, 3 etc.) corresponding to a spreadsheet decoding which are which.

That should not be difficult to do => happy to follow the suggestion.

Concerning the placebo, I am waiting for another few weeks to have a better view of the pets and pet owners. What I have in mind for now is the following: the researcher will provide one different number for each bottle and write the association in an excel file as planned, under a video taken by a thrid person. When a new participant wil need to receive several bottles for N rats I will tell the researcher that we need N bottles, he will use the excel file to send N bottles that all contain the same content (either placebo or treatment, I won't know). I will then send the bottles to the participant.

What about my objection to the public logbooks and the concern that this could influence results by allowing participants to see what others are experiencing. I am concerned that either of two things may happen:

- Participants will want their results to be similar to ones which other people posted
- Participants will want their results to be better than ones other people posted

Either case could cause severe problems in the data.

I am in complete agreement with this. The public logbooks will essentially break the blinding of the experiment. We seriously need to re-think them.

I wonder if we are not overcomplicating things and forgetting the forest for the first trees. I agree that those points are valid but I think that the open-science version has other positive aspects, if not greater positive aspects depending on what our utility function is:
- open-science largely improves things: many persons are following my experiment with mice at home and I have received great advice, both online and by PM or email.
- Also I think that mprize at home can be not only a pure experiment but also a great thing to follow collectively, to convey strong messages and interest on science and longevity to a wide audience.
Feedback on that is welcome.

Edited by AgeVivo, 16 September 2012 - 11:00 PM.

AgeVivo et-al:
I may have been over hasty with my last post. My apologies.
I'm glad you had the time to reply.

People I think we may be overcomplicating this.
The average pet rat/mouse lifespan is known.
There is no point in controls unless they are gettig exactly the same treatment/enviroment as the animals getting C60oo and are prefrably from the same litter.

I see no point in dissapointing people who love their pets and want the best for them. (Those getting placebo)
We already know that that is what we will be doing...

Even in a lab it would became obvious which animals where getting what and we would have to rely on the researchers to be impartial.

Just send everyone C60oo unless they actually want to do a 'double blind', placebo cotrolled study.

I agree with Logic and Agevivo. This is a community project, not a fully fledged lab research project. Maybe have 1 or 2 placebo and the rest C60oo. Maybe make it more intresting by varying the dose and dosing schedules in some sets of mice.

I think it is reasonable to assume that only about half the participants will run the full course and if there is nothing online to keep enthusiasm and interest up, I can see that completion percentage going down even further. Community projects tend to need constant reinforcement.

Even in a lab it would became obvious which animals where getting what and we would have to rely on the researchers to be impartial.

This would be true only if the C60 actually produces a positive effect.

We have one very small study which claimed that C60 in olive oil extended lifespan. As far as we know, that study is completely wrong and C60 could even cause premature death. This is why we want to attempt to replicate the study.

The only point of replicating this study is to get more data to determine whether or not the C60 is producing a positive, negative or no measurable effect at all.

The placebo may itself extend life, because it's olive oil and that effect was shown by the study we are trying to replicate. So again, without a placebo arm this is a pointless exercise.

If we give everyone C60, we will not get any usable data. Then there would be no point at all in this exercise, except to give free C60 to pet owners.

If you want to give your pet C60 go out and buy some and give it to your pet. Nothing is stopping you from doing that.

If you want to participate in a scientific experiment, and are willing to abide by the parameters of the study, and possibly give your pet plain olive oil, go ahead and sign up.

It's that simple.

My last post reminded me that the Baati study had three arms: Control, Placebo and C60.

I think we should add another arm which is Control: mice given neither olive oil nor C60.

If we can't find enough pet owners, that would be an issue. But the Control group has a really simple task: just stay in touch and let us know the starting ages and health, etc. of the mice, and when they die.

Again, public logbooks would be a really bad idea because the Control group might want to compete with the other groups and could have an incentive to claim that their mice lived as long as the others.

To my understanding, the only reason for having public logbooks is to increase participation and reduce dropouts. But I think there are other ways to handle this.

In my original suggested protocol posting, I suggested that there be a run-in period during which time the participants would have to regularly fill out surveys (and I suggested an online survey tool to facilitate this). The point of the run-in period is to verify that the participants are, in fact, participating. If they missed filling out the surveys, they'd be removed from the list and wouldn't get any samples. The hope is that the ones who participate for a fairly long run-in period (perhaps 2 months would be good) will be the ones less likely to drop out.

Another way we could create an incentive would be to publicly post the names of the participants, and then give them some kind of award certificate or something else tangible for completing the study. Participants who dropped out would be listed as "Failed to complete the study".

I'm sure other people can come up with more ideas which do not involve potentially unblinding the study and making any resulting data completely worthless.

Edited by smithx, 17 September 2012 - 05:55 AM.

The only point of replicating this study is to get more data to determine whether or not the C60 is producing a positive, negative or no measurable effect at all.

This is exactly the point I have been concetrating on. The difference between C60oo and nothing; not what effect a third substance may have.

The placebo may itself extend life, because it's olive oil and that effect was shown by the study we are trying to replicate. So again, without a placebo arm this is a pointless exercise.

The point of whether it is just the oo and not the C60oo that is making the difference is an important one. But one that I consider unecessary for the sake of keepig the experiment simple.

So the question becomes:
Do we want to kow what C60oo does compared to nothing, or do we want to know what C60oo does compared to plain oo?
My feeling is that we want to prioritise the difference between C60oo and nothing to keep things simple and exciting/fun, with a good chance of positive results and happy pet owners that will be interested in further experiments/research.
Remember that these people are not scientests ad are mre interested i keeping their pets alive and healthy for as long as they can.

I do however see your point Smithx, but feel that keeping things simple is more important. Exceptions could be made for those who are keen to do the kind of rigorous experiment, with controls, oo, and C60oo. (all 3)

The placebo may itself extend life, because it's olive oil and that effect was shown by the study we are trying to replicate. So again, without a placebo arm this is a pointless exercise.

The point of whether it is just the oo and not the C60oo that is making the difference is an important one. But one that I consider unecessary for the sake of keepig the experiment simple.

If we just do C60 and we seem to see mice living a bit longer, we won't know if it's the C60 or the olive oil that's doing it.

That's why I believe we really have to have C60, Olive oil and Control. Otherwise, any data we get will be useless and won't help people decide if they should take C60, which I think is really what we're trying to find out here.

I think the pros and cons have been largely described, over the last 24 hours, but also earlier. The truth is that there is not just one good solution. We should wait for a little more persons to create their logbook, the associated additional visibility would help shaping the experiment.. Please be patient: chating on a forum takes less time than organizing the experiment and having pet owners join us.

Edited by AgeVivo, 17 September 2012 - 07:31 PM.

Many points to respond to here, but I wanted to start with Logic. I am sorry you felt neglected. We are all thrilled that you and Maureen would like to participate. When we are trying something new across international borders, it takes a lot longer than we would like, and that can be frustrating. I urge you to please stick with us while everything is organized.

I also would be in favor of someone like Logic having controls, olive oil, and C-60 mice/rats in the same house. I am willing to take that chance, with the hope that we will also have many other C60oo, and placebo testers in isolation. I also trust that if we counsel the experimenters properly about the importance of placebo control AND they are active in maintaining an open log here at Longecity, we can (counter-intuitively) improve our chances of getting some meaningful results. This project has the dual purpose of engaging more people in the community AND producing new data with regards to C60oo. I think we should leverage the strength of our community and the openness of the project to produce good results.

We have 100 free samples of C60oo and we can bottle up 100 placebos. That produces a fairly nice sample size. We will have some problems. Some people will quit early. Perhaps some diversity in the base of experimenters will increase the odds of getting more of the rats and mice to make it to the end (happy and healthy of course , with an owner that cares about them). We might have a couple people like Logic that have the tenacity and capability of handling a few extra animals. I think we as a community can help him out if he is posting openly about how things are going. We can send reminders or helpful tips, things like that.

In the end, it is unlikely that our data will be published. I am a realist. But we could achieve something greater in community-based and supported science. We will have data that we can use and we will know how much validity it carries. Overall, it should be a benefit to the world at large, and for future gains in the battle against aging.

Edited by Mind, 17 September 2012 - 08:06 PM.

So how are we going to come to a consensus or decision here?

Niner and I have pointed out that having online logbooks is a really bad idea in that it has the very strong potential of completely corrupting the results.

I have pointed out that we really need three groups: control, olive oil and C60 olive oil to really replicate the experiment we are trying to replicate.

The question is: are we actually trying to get good data which people can use to inform their decision of whether or not to take this compound, or is this more of a community-building exercise and we don't care if there's no actionable data that results?

If the former, we need to be rigorous. If the latter, then we shouldn't pretend that we're really doing anything like a scientific experiment.

I have been hoping that we would be able to either replicate Baati's results or clearly show that they are subject to doubt. In either case, it would help me to make an informed decision as to whether to take C60 or not. I think this would be useful for lots of people (perhaps millions of people).

Or we can just have fun with mice and not bother about getting reliable data. How do we decide what it's going to be?

Niner and I have pointed out that having online logbooks is a really bad idea in that it has the very strong potential of completely corrupting the results.

My take on this, is that it is not really a bad idea. I think we can maintain good discipline through open discussion and monitoring of the experiment. My feeling is that peer pressure will produce discipline. Also, because no one will know what rats/mice are getting placebo or C60oo, AND that we will have rats of many different ages, it will not be so easy for experimenters to deduce what their mixture is in the first place, based on how their animals are doing. When people are doing the experiment and a particular rat is doing better/worse living longer/shorter we can publish reminders that they "shouldn't assume anything", that we "don't really know whether C60oo is going to extend lives", because Baati was such a small sample size. In my view more transparency=more discipline.

I side with open logs and trying our best to maintain a semblance of scientific discipline through the community effort. If we go the silent route, then we would need some other form of monitoring. We could try to set up protected anonymous log/forums with psuedonym names, which only a few people could view, but my feeling is that people would then just be more secretive in connecting with each other.

What if we were to set up google docs accounts, one for each pet owner? That way, we could monitor them, and each pet owner would see their own, but they wouldn't see other peoples' logs. I can appreciate the virtues of transparency, but when you're reading about other peoples' animals suddenly acting more frisky, but your own beloved pet is dragging, it seems like there's a big motivation to buy some C60-oo and dose up your little darling on the side. Maybe we could deal with that by encouraging people to drop out if they wanted to go that route, and assure them that we wouldn't be mad at them if they did. That would be a lot better than getting corrupted results.

the problem advantage is that it could encourage something of a competitive atmosphere, where different owners are trying to get their mice to be healthier or live longer.

I believe it would be an advantage for the quality of the study
I don't believe that working on a google doc, observed by 2 or 3 persons, is a sufficiently motivating motor for a good, well documented long term study: you would soon do it for yourself, make your own choices, stop documenting etc. Instead, if you do it for "others", you have a strong social existence through it that forces you and naturally drives you in doing the experiment as well as possible.

when you're reading about other peoples' animals suddenly acting more frisky, but your own beloved pet is dragging, it seems like there's a big motivation to buy some C60-oo and dose up your little darling on the side

This first seems to be a natural way but there is some blocking factor: it would mean that you believe that c60 works and that you are ready to prevent the experiment to detect that c60 works: this is bad for everyone's longevity.... bad for your longevity too!

Not only it is bad for everyone's longevity, you would take a dramatic reputation risk. Typically, if dosing with c60 end ups lengthening the hair (for example), you have a great risk to be discovered. Also, at any time we could ask to collect droppings and send them to us, for C60 analysis (for example; there are many things that can be asked). You? "Wanted -- Fraud against longevity"?

Edited by AgeVivo, 19 September 2012 - 09:53 PM.

Everyone posting publicly about what their mice are doing will absolutely create a competitive atmosphere in which the major activity will be trying to guess who has the placebo and who has the C60.

As a result of this, people who think they have the placebo will have a major incentive to get C60 and dose their mice with it if they want them to live longer.

The results will be useless if we follow this course.

(This idea of collecting droppings is silly because a) we have no idea if we could even do an analysis on them b) where's the money for that coming from? c) do we have data as to whether there would actually be much C60 in them anyway? d) why are we confident that someone who's fudging results would send us droppings from the same animals he's adulterating the diet of? So that idea is not realistic.)