Mining and protein folding: coins for progress

[Jacqul]

Hi,

Here is a quote from the wikipedia article regarding protein folding and Foldit:

"Protein structure prediction is important in several fields of science, including bioinformatics, molecular biology, and medicine. Successful identification of the structural configuration of natural proteins enables scientists to study and understand proteins better. This can lead toward the creation of novel proteins by design, advancements in the treatment of diseases, and the development of solutions for other real-world problems, such as invasive species, waste, and pollution.
The process by which living beings create the primary structure of proteins, protein biosynthesis, is reasonably well understood, as is the means by which proteins are encoded as DNA. Determining how the primary structure of a protein turns into a functioning three-dimensional structure—how the molecule "folds"—is more difficult; the general process is known, but protein structure prediction is computationally demanding."

Now we have a system that protects a protocol for transferring units of exchange i.e. bitcoins. As I understand it protects it by being computationally demanding and we now have chip circuits specifically built for this purpose and vast amounts of energy spent on performing these calculations.
Hopefully this will help drive the search for cheaper energy. It would be good if the computations themselves were a bit more useful (like Primecoin but actually useful).

I think a coin that gets "mined" by solving protein folding would be a perfect coin. Rewards can relate to complexity and stability of the structure. Solving natural protein combinations can be a built in test and potential initial distribution. Folding ("mining") could be by algorithm or by hand.
Solved protein architecture would be available through the block chain and effectively be open source.
When new folding solutions slow to a critical point POS based system of security could take over until the next level of complexity (for example interaction between different proteins according to their structure) is instituted.

This would not only revolutionize our system for communicating in terms of trade but it would also revolutionize medicine and manufacturing.


P.S. can I please start posting on the main forum

[maxcan]

I'm not an expert on how mining works but I think it's a bit different than protein folding simulation. The length of a protein doesn't necessarily dictate that it'll have more possible routes of folding, so increasing difficulty may be impossible.

I do like the idea (I use BOINC for protein folding and SETI), and would definitely support a coin that did more than waste electricity, e.g. NMC and XPM.

[Jacqul]

I agree, just translating the mining process as it is for Bitcoin into a system that does protein folding is a very simplified explanation about how it would have to work.
I expect the increased difficulty would happen after all the low hanging fruit (easily determined and already solved protein folds) are submitted. This would effectively look like a "pre-mine" and would have some of the same function but would also serve to start building up a library of proteins and to see how the system works.

To develop a system and coin like this would take someone very clever who knows both these systems well. Collaboration with the teams who are already involved with protein folding and its gamification (like Foldit) is also a good idea. I wish this was me, unfortunately it isn't but luckily there are people out there who are extremely intelligent (see Bitcoin) and maybe one of them will find this idea worthwhile.

[TiagoTiago]

How would folding a protein verify the validity of transactions?

[Jacqul]

Hi,

I don't know nearly enough about hashcash and bitcoin mining works and you ask an excellent question.
I have a feeling if someone has a good answer then they will also have a good idea of how it would be possible to implement a coin like this.
What I envisage (and it might be naive) is that the algorithms being used to solve protein folding at the moment could be adapted to have an inbuilt hashcash like function to validate computational effort spent on trying to determine a stable protein structure. Once a structure is "solved" this would be posted to the network and be part of the library.
So in my mind it would have 3 parts: the hopefully useful part of computation to solve protein structure, proven by published protein architecture of solved stable structures; the hashcash like part to verify and secure and prove work spent on solving (if the results themselves can not be used somehow); and the record of transactions which would make this system useful in transfer of units of wealth or worth and hence useful for trade.

[minerva]

How would folding a protein verify the validity of transactions?

how does prime number finding verify the validity of transactions?

why does even finding a hash even verify validity?

[Superdaan]

Curecoin is supposed to do this, though it has only been announced and not yet released.

[TiagoTiago]

How would folding a protein verify the validity of transactions?

how does prime number finding verify the validity of transactions?

why does even finding a hash even verify validity?

Sorry, i was very sleepy when i posted that.

Ok, the proof-of-work generation isn't the exact point where validation occurs. But having the proof of work not depend on the block you're trying to generate is an issue:

The hashes aren’t just to make life difficult for the miners. Because it takes so much computing power to find a hash that follows the rules, a would-be hacker trying to change the contents of a block would have to find a new hash that fits those rules again. Because he’s trying to rewrite history he not only has to find a new hash, he has to find it faster than the legitimate miners can, which would require a tremendous amount of computing power and confers very little benefit. The more miners there are, the higher the difficulty goes. The higher the difficulty goes, the harder it is to commit fraud on the Bitcoin network.

[Jacqul]

Curecoin is supposed to do this, though it has only been announced and not yet released.

Thanks Superdaan, I just read the curecoin announcement with a bit of disbelief (don't know how I missed that one). Seems I am not the first to have this idea at all! I'm slightly disappointed but also happy that there are people working on something like this already. I hope they get it right. It seems there are still some problems to iron out.

[selenahill21]

Fascinating! I have imagined that game can use the computing power of the players’ computers but this is a new approach to solving scientific problems. Since 1996, treating HIV and AIDS has not transformed much. The same narcotics on basically the same routines are prescribed. How, exactly, the enzymes and proteins of HIV and AIDS work together have been unknown. Researchers have, for the last 10 years, been trying to figure out this challenge. Once scientists decided to ask players for their help, the protein structure was untangled in just three weeks.

[ranlo]

Curecoin is supposed to do this, though it has only been announced and not yet released.

Curecoin (CURE) has been out for a long time. It uses a special method, though. Basically you use FAH (Folding At Home) and x coins are spread among people that fold, based on their performance compared to the overall (so if you were 50% of the network, you'd get 50% of the generated coins).

While the folding itself doesn't give coins, this method helps facilitate a medium to do the same.

And FWIW, FAH is pretty stable and has been for months.

[sl@ppy]

This is the future for many in getting health care.  Digital currencies will work their way into health and those that sit at home all day playing will be able to see doctors

[Bizmark13]

Curecoin is the one that first came to mind when I read this thread. There is also a similar coin called Gridcoin. Gridcoin uses BOINC while Curecoin uses Folding@home. There are many BOINC projects, and one of them is called Rosetta@home which also deals with protein folding. In terms of market cap, Curecoin is currently considered to be the more successful coin of the two.

It should be noted however that neither Curecoin nor Gridcoin actually pay you to perform protein structure verification directly. Instead, you submit shares to either BOINC projects or Folding@home, respectively. Then the miner measures your contribution to these projects and in comparison with the average contribution of the whole network, calculates your reward accordingly. Hence the "folding" bit is still centralized (unlike Primecoin and Riecoin where all the algorithms and calculations required to perform research are actually embedded into the protocol itself in a fully decentralized manner).

As a transhumanist and Extropian, I always wondered if it might be possible to decentralize the process of calculations required to perform and verify protein folding and other complex scientific experiments. That would definitely make for an interesting coin.

[TrailingComet]

I think these arte great ideas that I hope will be developed further in coming iterations
Fact is all that computational power should be put to its most productive use