Rosetta@home

Rosetta@home
Name	Rosetta@home
Launched	2004
Developer	University of Washington Baker laboratory
Platform	BOINC
Status	Active

Rosetta@home Rosetta@home is a volunteer distributed computing project that leverages spare cycles on personal computers to perform large-scale protein folding and protein design computations supporting biomedical research. It connects volunteers worldwide with research teams at institutions such as the University of Washington, enabling contributions to studies in structural biology, computational chemistry, and virology. The project has collaborated with laboratories across institutions including Harvard University, Broad Institute, Wyss Institute, and Cold Spring Harbor Laboratory.

Overview

The project uses the BOINC platform developed at University of California, Berkeley to distribute workloads to volunteer hosts, coordinating tasks via servers hosted by the University of Washington Department of Biochemistry. It applies the Rosetta suite of algorithms originally created by David Baker and colleagues to predict three-dimensional structures from amino acid sequences and to design novel proteins. Target areas have included structure prediction for proteins from organisms studied at European Molecular Biology Laboratory, antibody modeling pertinent to work at National Institutes of Health, and design challenges encountered by teams at Scripps Research and Massachusetts Institute of Technology.

History and Development

The initiative originated from research in the Baker laboratory at the University of Washington during the early 2000s, building on algorithmic foundations developed by researchers including David Baker, Jeffrey Skolnick, and collaborators at University of Illinois Urbana–Champaign and Stanford University. Early deployments paralleled the rise of distributed projects such as SETI@home and were influenced by infrastructure work from BOINC developers like David Anderson. Milestones include contributions to the Critical Assessment of Structure Prediction (CASP) experiments, collaborations with the Protein Data Bank, and integration with computational efforts at institutions such as Carnegie Mellon University and University of Toronto.

Software and Distributed Computing Infrastructure

Work distribution relies on the BOINC middleware, with server farms maintained by university IT groups and computational workflows orchestrated by researchers affiliated with Seattle Biomed and the Allen Institute. The client software executes the Rosetta suite, which comprises modules for comparative modeling, ab initio prediction, docking, and de novo design—techniques refined alongside groups at University of California, San Diego, University of Cambridge, and Imperial College London. Results are aggregated and analyzed using high-performance computing resources at centers such as Argonne National Laboratory and Lawrence Berkeley National Laboratory and visualized with tools developed by teams at UCSF and European Bioinformatics Institute.

Scientific Contributions and Applications

Computational outputs have informed experimental work across multiple institutions, aiding structure determinations deposited in the Protein Data Bank and guiding wet-lab validation at facilities including Brookhaven National Laboratory and Max Planck Institute for Biochemistry. The project contributed models used in research on viral proteins studied at Centers for Disease Control and Prevention collaborations and to design efforts for therapeutics explored by groups at Genentech, Regeneron Pharmaceuticals, and the National Institute of Allergy and Infectious Diseases. Achievements include participation in CASP rounds alongside entrants from University of California, Irvine and successes in protein‒protein docking evaluated at meetings of the American Chemical Society and Biophysical Society. Designed proteins and binders have been co-developed with laboratories at Yale University, Princeton University, and Johns Hopkins University.

Participation and Community

The volunteer base spans individuals affiliated with academic institutions such as Massachusetts Institute of Technology, University of Oxford, and Peking University, as well as hobbyist groups and citizen scientists organized through online forums and community hubs tied to entities like the Open Science Grid. Community coordination draws on experience from large volunteer projects including Folding@home and World Community Grid, with competitions, team-based credit systems, and recognition echoed from practices at XSEDE and regional computing consortia. Outreach has involved presentations at conferences hosted by the Gordon Research Conferences and workshops at Cold Spring Harbor Laboratory.

Security, Privacy, and Ethical Considerations

Operational security encompasses measures implemented by the hosting institution, leveraging policies aligned with National Science Foundation guidelines and IT standards practiced within universities such as University of Washington and University of California, Berkeley. Privacy practices address volunteer data handled under institutional review frameworks comparable to processes at Harvard Medical School and Stanford Medicine, while ethical oversight references norms promoted by organizations like the Wellcome Trust and Howard Hughes Medical Institute. The project has navigated issues similar to those encountered by distributed initiatives at Los Alamos National Laboratory and during debates at forums convened by the IEEE and ACM about volunteer computing ethics.

Category:Volunteer computing projects Category:Computational biology Category:University of Washington