I just rejoined after being away for awhile we working on a different project but they had estimated the computing time on 'one' PIII @ 1ghz would of taken almost 28.000 yrs at 24/7 and 100% computing power/speed. I however rejoined after recieving a 'spam' mailing in my yahoo mail which reminded me of the grid. However I have not seen even one reference to it publically in Canada. Perhaps more awareness would cut the puting time considerablly. I am also looking for a Team to join with in Canada preferably but anywhere is fine.

http://www.worldcommunitygrid.org/stat/global.html

So if there are 50,000 proteins, and a total of 21,553 members (at the time of this writing) then what does that say about the level of redundency?
Obviously there must be more to it...Perhaps each protein gets analyzed a number of different ways, or perhaps there is actually a lot more than just 50,000 proteins being tested here.

I agree. It would seem improbable that they could assign each protein a unique range of values when the essence of the project is to start with a basic unit and evolve it into various conceivable shapes.

hmmm...Perhaps it is more evolutionary than we think. If it is in essence, evolving, via a sequence of random numbers/algorithms - then it would make sense that each run on each machine would give it the maximum probability of what it is most likely to evolve into.

Certainly there could be a number of environmental variables that they could use as differentiating each individual protein, but this is not written anywhere that I have seen.

No·MAD

Viktors 3 Dec 2004 on Screen Saver and Random Seed for Protein Folding:
http://www.worldcommunitygrid.org/forums/wcg/viewthread?thread=783

Since we are returning the structures (in some form or another) that we have created through folding together with the score for that structure, my personal guess is that the Institute for Systems Biology will do some additional computing of their own for likely structures. For all I know we (the Project as a whole) may be creating and scoring hundreds of thousands of possible configurations for each protein.

We will be getting some more information when the Human Proteome Folding Project staff at the Institute for Systems Biology start their posting on the progress of the Project. Did you catch that post from the Project Manager?
http://www.worldcommunitygrid.org/forums/wcg/viewthread?thread=908

My idea is that each protein is sent out many times, but the Work Unit is assigned a different set of seed numbers used in generating trial configurations. I do not think that there is large scale duplication of effort. Of course, there may be a number of cases in which identical configurations are generated starting from different seed numbers. Anyway, this is just guessing. But I am sure that a great deal of thought has been put into maximizing the return from computer time.