one more thing,
the place to start understanding the project is suposed
to be the human proteome folding web site that is a dirrect link
off of
[url=http://systemsbiology.org]

does this page make sense? should I revise it?
what parts do and don't make sense.

Hello barney15c,

Most of your questions are not really responsive to Dr. Bonneau's call for questions, so I will try to answer them for you. Dr. Bonneau is trying to improve the ISB web page at http://www.systemsbiology.net/Default.aspx?pagename=humanproteome which explains the Human Proteome Folding Project. He needs to know what questions it leaves unanswered and which explanations in it need clarification.

1) Will there be a second phase to the HPF project.
IE those proteins that have been successfully folded in this project will be further folded to a greater degree?
2) Will an announcement be made as to what the next WGC project is going to be, be made before the completion of this one, so as not to lose the inertia of participants joining the WGC. I fear a gap between projects would mean that we would be moving back a step as people would lose interest in waiting for the next project.

My answer is unofficial since I am a volunteer forum moderator. These 2 questions are related to each other. If you read the post ‘Re: What's next?’ by bbover3 at http://www.worldcommunitygrid.org/forums/wcg/viewthread?thread=2612#16941 then you will see that it is taking us time to acquire projects in other fields that meet our requirements for benefiting humanity. Scientists do not start planning projects that require vast amounts of computer processing until they are aware that they can acquire the needed resources. So it seems nearly certain that the ISB and WCG will continue to collaborate by folding some selected proteins believed to have special significance using the more accurate high-resolution method that Rosetta can use. Currently we are folding proteins using a much faster, but less accurate, low-resolution method.

This ensures that there will be no 'dead time' when we do not have a project to run. The amount of time spent doing this depends very much on the perceived importance of the results as compared with any competing projects so the people that make these decisions or conduct negotiations will probably prefer to only announce decisions after they are made rather than discuss the arm-wrestling used to reach a decision.

Since it takes time to 'board' a program for the WCG, announcements of new projects will be made well before they start running on our computers. This has always been promised on the WCG web site and the promises have been repeated.

3) Any progress on a linux version of the rosetta application ?

This was explained around the end of 2004 in http://www.worldcommunitygrid.org/forums/wcg/viewthread?thread=1246
Rosetta began as a Fortran program running on Linux. We are currently running a version converted to C++ on the Windows operating system. A number of functions we are not using are stripped out to reduce the memory footprint of Rosetta. Because of its Fortran origins, Rosetta has a lot of statically allocated arrays, which accounts for its large virtual memory size. The actual working set of memory pages was about 75 MB in size in November 2004 and was later reduced to about 25 MB. So there is no special Linux problem with the application. The WCG has started out using a Windows Agent and is using the experience gained to design an Agent to run in Linux. Whether or not we shall ever run some version of Rosetta in Linux is not a decision that is affected much by programming constraints.

4) Does the WCG have any plans to increase awareness (publicity campaign etc) to bring more people on board to participate in crunching for this/next project (whatever that may be)?

I know about as much as you at this point, but perhaps a little more. There is an opinion that we should start out by publicizing results achieved rather than by cheer-leading results to be achieved. Opinions are always subject to revision and change. The long-term intention is to expand to many times our current size. Having a large number of researchers pushing projects in many fields would probably raise the interest in conducting a large-scale publicity campaign to expand the World Community Grid. In previous posts I have stated my opinion that these distributed computer projects will eventually be as widespread as bake sales in schools and that children will grow up contributing their spare computer time and taking it for granted.

Reiterating, Dr. Bonneau is asking what needs improving in http://www.systemsbiology.net/Default.aspx?pagename=humanproteome

What questions do you need to ask after reading it?

mycrofth

1. how do we tell if a prediction is good:

we have a confidence function that takes as inputs
the protein length the degree to which the rosetta program converged
and then outputs a confidence value that one can use to judge the degree to which one should trust the fold prediction.

2. Will there be a publicly available summary of verified structures?

this is difficult to do until the post processing is done but we'll
have overall stats as the results roll in. Some of these stats
will be a bit technical, but I can try to put something up somewhere
that says how many proteins we've predicted with > 80% confidence.
This should be a satisfyingly large number.

Yes, the HPF2 project is one of the proposals which has been submitted and is going through the review process. The last step will be the advisory board. The possible start date is still several months away. And, there is still plenty to process on HPF1.

I'm curious to know the status of the Human Proteome Folding project. When do you think it will be finished? What's coming next?