Real Science or just another Video Game ?

Case number:699969-989556
Topic:General
Opened by:itskimo
Status:Closed
Type:Question
Opened on:Friday, April 15, 2011 - 15:38
Last modified:Tuesday, April 19, 2011 - 22:02

somehow when i started Foldit i thought it had something to do with solving science problums. I was wrong. Guide puzzles are not science, they already have the guide. de novo puzzles are not science, spegetti solutions have no value to science. interface puzzles are usualy just frustrating, so where is the science?
If all these puzzles are suppose to teach us how to fold why are there no advanced tutorials?
If this is realy for Science, what and where is the feedback on what we have done to advance the science of Foldit? what original solution in the past 6 months has made a contrabution to science. I for one would like to know what it was and where is it?

(Fri, 04/15/2011 - 15:38  |  9 comments)


Joined: 08/24/2010

whoa kimo kimo calm down :)

1. Advanced tutorials will eventually be developed for the wikis
2. they think chaperones may use a "fusing" technology in combination with friction (heat) to aid protein folding
3. molecular weight is vital in structure prediction

..the breakthroughs are coming :)

B_2's picture
User offline. Last seen 4 years 16 weeks ago. Offline
Joined: 11/29/2008
Groups: None

None of which seems to answer the OP's question.

1. Wikis are useless - people won't read wikis, the tutorials need to be hands on, interactive, with very detailed and non-biologist-understandable explanations of exactly WHY we should be doing what is being demonstrated. The current alignment and exploration tutorials are good examples of useless tutorials, since they do not explain nearly enough in plain language what we're supposed to do and WHY. They may explain the mechanics of using the tool, but that's it.

2. What's a "chaperone" in this context?

3. Huh? Molecular weight? Never heard that term associated with foldit so far.

Kimo is correct - without proper context and feedback, what we are doing is simply a video game for the highest scores, regardless of whether what we're producing makes sense.

Pletsch (not verified)
Pletsch's picture
Groups: None

I think maybe there is more expectation here than should be, I don't see our role to be that of amateur scientists hell bent on finding a big solution to some mysterious problem. We are more like parts in the process trying to help out to furnish large amounts of results in order to assist the research of a "real" scientist. The "work" we do here is not the role of scientific research, it's more akin to the role of scientific equipment. Because the equipment is not sophisticated enough to complete tasks quickly and return amazing results there is an opportunity for us to contribute meaningful results in a way which may help to create better machines at some point. What we are doing is essentially providing data that can be analyzed so that better computing programs can be devised to replace us at some point. I don't see this as a bad thing, only a fun source of recreation that may aid in creating useful tools someday. The chance of finding a real solution, or getting lucky and making a breakthrough are enticing, and attractive to many, but ultimately I do think it's more about the game, contributing, and seeing how people stack up vs. machines, and to allow those programming the machines to perhaps learn something about what people do differently than existing programs which may help to replace us someday. Of course the opportunity to discover something really huge is there, though I think it's an extremely slim chance, and the expectation of valid real world results is a bit out of line with the tasks which this program tackles. I may be wrong, but that's just my two cents.

The game does accomplish real science, just not in the way many people like to think imho.

zoran's picture
User offline. Last seen 6 years 28 weeks ago. Offline
Joined: 11/10/2007
Groups: Window Group

Actually, in fact we currently have 3 papers in submission, and a number more pending. here's the summary:

1. a protein that has not been solved for 15 years has recently been resolved by foldit players and confirmed by x-ray crystalography. the paper is currently submitted to science. The group that produced the solution is on the author list.

2. analysis of macros used by fodlit players lead to a comparison of those macros against the state of the art algorithms. we found that some of the frequently used macros outperform standard proteomics algorithms, and are actually a variant of fast relax protocol that is currently not published but is known to outperform other methods. the paper is in preparation to be sent to PNAS.

3. a new enzyme has been developed AND confirmed in the lab that has improved over other known enzymes focusing on Diels-Alder reactions. In this case the scientists used the novel backbone configuration from the foldit players, and redesigned the sidechains. the paper is in preparation.

So in summary, over the 2 years of foldit life, we have both shown that foldit players can outperform known computatonal methods, discovered the structure of an unsolved protein, discovered novel automated methods for protein relaxation, and designed new proteins that are confirmed in the laboratory. I don't think any individual lab in the world could boast with these kinds of outcomes in just 2 years. Of course, i only expect us to do even greater things in the future.

So no need to lower the expectations. Foldit player should be proud of what they have accomplished. In my view, the most important outcome of this game is that we may have tripled of quadrupled the numer of world experts in protein science in two years.

Joined: 06/17/2010

Yay! We rulz :)
Can`t wait to see links to all that papers :)

Joined: 09/16/2010
Groups: None

Yay, And I'm waiting for a link to the source code too.

Brick's picture
User offline. Last seen 4 weeks 1 day ago. Offline
Joined: 07/15/2008
Groups: Beta Folders

How can you have scientific results that can be considered valid with an open source science application?

People can make the result be anything they want it to be, maybe not to trash the science on purpose, but certainly to game the scoreboard.

Joined: 06/17/2010

Oh cmon Brick. Even when it will be OS counting scores have to be sent via valid (closed) client using its scoring database, cypher keys, ssl etc. I see no other way. Fold sent to central can be easily validated on theirs side, so I see no problems.
Only good thing can happen after source will be opened is faster, stable and more customizable client :)

Joined: 11/21/2008
Groups: None

While I appreciate the comments regarding the aims of fold.it, I agree with B-2 on tutorials. I would like to see 2 Intros for partial threading and (when the current problems are addressed) also for Exploration. Clear hands-on instructions, similar to the current Intros would also help to keep new players, the lose of which was mentioned in chat yesterday. If I was starting now I would probably give up after a short time, and I have read the wikis many times times.

Sitemap

Developed by: UW Center for Game Science, UW Institute for Protein Design, Northeastern University, Vanderbilt University Meiler Lab, UC Davis
Supported by: DARPA, NSF, NIH, HHMI, Amazon, Microsoft, Adobe, RosettaCommons