See the bottom of this page for a description of the icons on the page.

The client currently participates in one or more of the following projects: an exhaustive regression analysis to identify the specific factors that ease the suffering of chemotherapy recipients, a gene expression analysis and a study of the molecular dynamics of protein folding to create greater understanding of the behavior of cancer cells and how they interact with potential new treatments, and an exhaustive regression analysis of clinical trial data for Amarillo Biosciences, Inc. for a treatment of fibromyalgia syndrome.

On November 16, 2004, the project announced that 400 compounds from the LigandFit phase have been synthesized and tested in a lab. Between 2-4% of them show activity. Typically less than 0.1% of compounds from computer-simulated screening show activity. The next step for the successful compounds "is to persuade pharmaceutical or biotechnology companies to take up these experimentally verified hits and to test them further."

The project uses the grid.org computing platform. See the grid.org platform information for the latest version of the grid.org client. The grid.org client is available for Windows.

Join a discussion group about grid.org's projects.

On October 21, 2002, the project announced that it achieved its initial goal: to "simulate the folding dynamics of proteins and make quantitative predictions for how [protein folding] works. This has been a 'holy grail' of computational biology." Specifically, the project simulated the folding of a man-made chain of 23 amino acids called BBA5. The simulation's measurements and folding-time matched physical measurements and folding-time of the protein observed in a lab. The achievement was published in the November, 2002, edition of Nature, in an article titled "Absolute comparison of simulated and experimental protein-folding dynamics." In October and November, 2004, the project's lead developer and research scientist, Guha Jayachandran, won the top award at the 2004 Biomedical Computation at Stanford (bcats) conference, and Vijay Pande has been named a 2004 Technovator.

On January 10, 2003, the project broke the 90,000 active CPU mark. On October 10, 2003, it broke the 120,000 active CPU mark. On April 24, 2004, it broke the 150,000 active CPU mark. On September 9, 2004, it broke the 1,000,000 total CPUs and 100 TFLOPs marks.

On November 28, 2001, the project began studies (110 and 5100) on Alzheimer's Disease. On May 8, 2002, the project began studies (503 and 504) on Huntington's Disease, a disease similar to Alzheimer's Disease in that it is caused by protein misfolding. See a list of all of the project's current studies. In June, 2002, the project started using the Gromacs molecular dynamics modeling software. This software allows the project to simulate larger proteins more accurately and much more quickly than the previous software. As of August 12, 2002, the Gromacs software has successfully modeled one of the simpler molecules, project 902, and shows a lot of promise for success with larger molecules. On March 27, 2003, the project began studying several new types or work units, including NTL9 (a protein with both beta and alpha structure) and 1PRB (a three-helix bundle). The project also increased the scope of its trpzip investigation to include a fourth variant trpzip4. On April 23, 2004, it published the first results from its Gromacs core. On August 20, 2004, the project began its P130x projects: these projects have work units much larger than normal, and require users to have hundreds of MB of RAM and be able to transfer 5 MB of data for each result, but the work units have 50% more value in stats. On October 12, 2004, the project released a new AMBER (Assisted Model Building with Energy Refinement) core which will allow it to do things that it can't with its existing Tinker and Gromacs cores. See the AMBER core FAQ.

Version 5.02 of some of the Windows, Linux and Mac OS X clients is available as of August 20, 2004. Version 4.00 of the other clients is available as of December 19, 2003. Note the 2.x clients have a compatibility problem, and you should upgrade to the latest client if you are using an old version. Beta version 5.03 pre 9 is available for Windows.

Students and teachers should see the Education@Home section of the site.

See the research papers and articles published from the results of this project.

See Professor Pande's February 19, 2004, talk at Xerox PARC: Folding@Home: Can a grid of 100,000 CPUs tackle fundamental barriers in molecular simulation? Windows Media Player is required to view the video.

See independently-maintained hourly team and user stats for this project at statsman.org.

See an independent support site for helpful hints about participating in this project.

See the original Folding@Home project.

Join a discussion group about this project.

Join a discussion group about the original project.

Join an independent discussion group about ab-initio protein folding (where the folding is simulated completely in software)

The project completed Phase I on May 21, 2003 (in this phase, almost 60,000 computers completed 1,400 years of computing to process over 9 million tasks).

"[Phase I] has ably demonstrated that with such massive computational abilities, researchers can utilize intensive approaches to identify drug candidates that succumb to resistance mutations and those that are more resilient to such mutations. An early lead developed during Phase I, TL-3, has been shown to be promising against the drug resistant strains that have arisen from the currently approved HIV Protease inhibitors. The characteristics of TL-3 have been born out by the FAAH computational work."

The project began Phase II on May 21, 2003. Note that for Phase II, the project transfered completely to The Scripps Research Institute (and became a completely non-profit project) and Entropia is no longer involved. You will need to reregister and download a new client to participate in Phase II. You can learn more about the project through articles it published on January 19 and January 26, 2004:
The Resistance Part I: From Petri Dishes to Population Dynamics
The Resistance Part II: Fighting HIV Resistance At Home and in the Laboratory

The client is currently available for Windows. It is best for users with permanent Internet connections. The client will eventually be available for Mac OS X and Linux.

The Beta version 0.7.1 of the client is available as of January 8, 2002. This version of the client has support for users behind proxy-based firewalls, uses your CPU and network bandwidth more efficiently, and has many other new features. The client will run on any platform that will run Java 2. This project is better suited for users with full-time Internet connections because a work unit completes in only a few minutes on my Pentium II 350 and the client doesn't buffer work units. Note: as of November 8, 2004, the project is moving to a new data center environment. Users should edit their agent.cfg files and change the "Coordinator-Address" field to the new value of dctp.ubero.net.

View the project's discussion forum.

The graphical client can run minimized or (on Windows) in the system tray. Currently it does not allow you to select which protein target(s) to work on: future versions will allow this. It does not require a continuous Internet connection: it allows you to set the number of work units it will cache so that you can cache more work if you connect to the Internet infrequently. Note that if you pause or restart the client, you will lose all of the conformers you've generated for the current work unit. The client supports users behind firewalls. Version 2.0 of the client is available for Windows, Mac OS X, Linux and Solaris as of September 18, 2003. See a press release about the significant updates in this version.

Join the D²OL discussion forum.

See a white paper about this process: The Virtual Laboratory: Enabling Molecular Modeling, and a PhD thesis. The biology collaborator working on this project has already designed a drug to counteract an ECE enzyme involved in heart stroke, and is currently using the Virtual Laboratory to study liver cancer.

See another white paper by the project coordinators: Grids and Grid Technologies for Wide-Area Distributed Computing.

Note: this project is currently only for Linux/Unix users. Windows is not supported yet.

The project tested its 1 billionth conformer in July, 2004.

The project uses the D²OL software client. On Windows the client runs in the system tray, but in all versions it also has a graphical interface and a command-line interface to allow you to watch what it's doing. It does not require a continuous Internet connection: it allows you to set the number of work units it will cache so that you can cache more work if you connect to the Internet infrequently. Note that if you pause or restart the client, you will lose all of the conformers you've generated for the current work unit. The client supports users behind firewalls. Version 2.0 of the client is available for Windows, Mac OS X, Linux and Solaris as of September 18, 2003 (see a press release about the significant updates in this version).

Join the TSC discussion forum.

On June 30, 2003, the project announced that it successfully found some potential cancer-inhibiting drugs. After testing over 500 million molecules, it produced a set of candidates. 39 candidates were tested in a laboratory, and 7 of those (20%) showed the desired anti-cancer properties. Only 2-3% were expected to show the desired properties. On November 12, 2003, the project announced that it has found growth inhibitors for 5 anti-cancer protein targets. "42 of the 200 molecules tested [so far] showed the desired anti-cancer properties." On March 22, 2004, the project announced that it "has found more molecules which inhibit the growth of cancer cells," and has now discovered "growth inhibitors for 7 anti-cancer protein targets." On April 16, 2004, the project announced that it has found 18 molecules which inhibit the growth of HIV protease, an AIDS protein. More of its cancer-fighting drugs were confirmed to be effective in laboratory tests as of July 2, 2004.

The Cancer and Proteome projects are active as of August 8, 2002. The Bioterrorism Antidotes project is active as of October 15, 2002. A new methodology project is active as of October 26, 2002 (methodology projects improve the performance and accuracy of the Think software). A new protein, Tyrosine phosphatase (PTP1B) was added to the methodology project on February 19, 2003. The Respiratory Disease (starting with SARS) and Multiple Sclerosis projects are active as of May 7, 2003. A second SARS target was begun on May 29, 2003. The HIV project is active as of July 1, 2003. The Malaria project is active as of May 10, 2004.

The client supports users behind firewalls. It allows you to easily opt in and out of any of the drug-finding projects, so you can participate in only the projects that interest you. And it allows you to buffer work units so that you can work off-line. Note that if you have work units for a protein query which has been closed, those work units will not be used or counted if you return them more than 30 days after the query has ended. Version 1.25g of the client for Windows and Linux is available as of November 12, 2004.

Join a discussion forum about the project.

Note that this project is in its alpha test phase. Credits and results may be reset or deleted at any time. Note also that the project is on hold as of November 1, 2004, while it submits a grant for more funding, installs a new project server, and migrates its applications to BOINC 4.x.

The top 5 predictions for targets T0196 and T0197 are available as of July 1, 2004. You can see them in the project's gallery.

The project uses the BOINC computing platform. See the BOINC platform information for the latest version of the BOINC client. The BOINC client is available for Windows, Linux and Mac OSX. Note that the client needs 150 MB of disk space and 50 MB of RAM to run. You may need to adjust your available disk and memory percentages to allow the client to use these amounts. The project server was upgraded to BOINC 3.19 on June 27, 2004.

Join a discussion forum about the project.

The project was presented at the 2004 Biomedical Computation at Stanford (bcats) conference on October 16, 2004.

To participate in the project, follow the installation instructions (text-only version) and send email to charles.parnot@stanford.edu so that he can contact you with project updates.

The project uses the World Community Grid or grid.org computing platforms. Volunteers who are not already registered with grid.org are encouraged to participate through World Community Grid. See World Community Grid and grid.org platform information for the latest version of these platform clients. The clients are currently available for Windows. Work units for the project time out after 4 weeks. Client systems must have a minimum of 256 MB of RAM in order to participate in the project. The latest version of the Rosetta core was released on November 29, 2004.

Join a discussion forum about World Community Grid's projects.

Join a discussion forum about grid.org's projects.