This is an old revision of this page, as edited by MoonFlute (talk | contribs) at 07:48, 5 August 2006. The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.
Revision as of 07:48, 5 August 2006 by MoonFlute (talk | contribs)(diff) ← Previous revision | Latest revision (diff) | Newer revision → (diff)
The United Devices Cancer Research Project is one of several distributed computing projects that have been operated on the grid.org website by United Devices.
Cancer Research Project
The United Devices Cancer Research Project, which began in 2001, is seeking possible drugs for the treatment of cancer using distributed computer power. There are around 150,000 users in the United States and 170,000 in Europe along with hundreds of thousands more in other parts of the world.
The project is an alliance of several companies and organisations:
- United Devices Inc.
- National Foundation for Cancer Research
- University of Oxford Department of Chemistry
- Donors of molecular research
United Devices released the cancer research screensaver under the principle of utilising spare computing power. The program, which can also be set to run continually, uses "virtual screening" to find possible interactions between molecules and target proteins, i.e. a drug. These molecules (ligands) are sent to the host computer's UD Agent. When these molecules dock successfully with a target protein this interaction is scored for further investigation.
Phase 1 has been completed and Phase 2 is underway. In Phase 1, over 3 billion drug-like molecules were tested against 12 proteins which are known as suitable targets for anti-cancer drugs. It used the "THINK" software for the simulation of the molecular interactions.
Phase 2, using the "LigandFit" software developed by Accelrys to model interactions, seeks to refine the Phase 1 data to produce a more manageable list of drug candidates for testing that will require experimental collaborators, including some from industry.
Human Proteome Folding Project
The Human Proteome Folding Project was announced on November 16, 2004 as a joint partnership between United Devices and IBM. In order to accommodate the large number of expected participants, the project would be operated simultaneously on both the grid.org servers and on IBM's World Community Grid servers.
It makes use of the "Rosetta" software to predict the structure of human proteins in order to help predict the function of proteins. This information may someday be used to help cure a variety of diseases and genetic defects.
Past projects
Anthrax Project
The Anthrax Research Project was a part of the United Devices "Patriot Grid" initiative to fight biological terrorism. It made use of the "LigandFit" software (that had already been used by phase 2 of the Cancer Research project), but with a specialized set of target molecules that targeted the advanced stages of anthrax bacterial infection.
The project was operated from January 22, 2002 until February 14, 2002 and ended after a total of 3.57 billion molecules had finished screening. The results of the research project were transmitted to biological scientists in order to finish the screening of the computational simulations.
The partners of the project included Oxford University.
Smallpox Project
The Smallpox Research Grid was a part of United Devices "Patriot Grid" initiative to fight biological terrorism. It made use of the "LigandFit" software (that had already been used by phase 2 of the Cancer Research project), but with a specialized set of target molecules that targeted the smallpox virus.
The partners of the project included University of Oxford, the University of Western Ontario, Memorial Sloan Kettering Cancer Research Center, Essex University, Evotec OAI, Accelrys, and IBM.
HMMER Project
The HMMER Genetic Research project made use of the Hidden Markov model to search for patterns in genetic DNA sequences.
Webload Project
The Web Performance Testing project was operated as a commercial opportunity with select web hosting providers in order to help them test the scalability of their server infrastructures under periods of high-demand.
Wasting its members' resource
According to Grid Forum's announcement since July 03 2006, one of its two active projects "The Human Proteome Folding Project" has been completed but left running, wasting its member's processing resource.
"Rosetta Job - There are no more new batches of data available to us to work on. The WUs that are currently active have all been previously processed so the work is redundant."
Apparently, Robby, an UD employee believes that members would like to have the redundant project running because "Maybe some members think the Rosetta screensaver is cool." This has caused some controversies in Grid Forum, since most of its member were home users who would like to contribute to some useful scientific researches.
According to member cdfalcon24, by July 17 2006 "grid.org has made the decision to leave the Human Protomene Folding project running, even though it is no longer contributing to scientific research. All work on the HPF project is now for points only. Here are the statistics for how much effort was put into the HPF yesterday: 32 years 320 days 9 hours 56 minutes 38 seconds; 7,434,100 points ; 16,097 results. This is an enormous amount of computing power that is being wasted when it could be put into the cancer project. The cancer project is up and running, performing meaningful research that may lead to a cure for cancer."''
Despite its member suggestion to shut down the wasteful project, the Grid.org staffs have refused to listen:
"Speaking of Robby, I PMed him twice asking him to read this thread, and he has not given any response. It would appear that grid.org's strategy is to:
1. Never update the client, instead leaving it incapable of utilizing 64bit processors, dual core/dual processors, extra crunching time by downloading a new WU shortly before finishing the current one, extra crunching time by running as service, etc.
2. Crunch meaningless data, diverting processor power from the real work.
3. Fail to inform users that they are no longer crunching meaningful data.
4. Ignore users who have voiced their opposition to this behavior.
5. Use forum "moderators" to censor any suggestions that users move to another project that is actually doing meaningful work with an updated client.
The one thing that grid.org has going for them is the cancer project, offering us a hope for a cure for cancer. But this hope is overshadowed by the mismanagement of grid.org. If grid.org does not care enough about the cancer project to divert resources to it from the now-finished HPF project, then how much confidence do we have that the cancer project is doing something meaningful?
How to use your processing resource for active cancer project only
Go to the home page, log into the "member services" page, click on "My profile manager" on the left panel then choose the Profiles in the three choices "Devices | Profiles | Statistics". Click on the profile you want to change and uncheck the check box for "Human Proteome Folding Project", then click save.
See also
External links
- BBC News report
- Grid.org
- The National Foundation for Cancer Research
- United Devices Inc.
- Department of Chemistry at the University of Oxford
- Evotec OAI - drug R&D company