We propose to add 1,760 processor cores to the existing shared computing cluster at California Institute for Quantitative Biosciences (QB3) at University of California, San Francisco (UCSF). The upgraded Linux-based cluster of 6,338 processor cores will provide supercomputer class performance at a fraction of the cost, approximately doubling our current computing power. The cluster will be used by 12 Major User research groups and others, for the total of 87 research groups with 455 registered users, of which approximately 400 are graduate and postdoctoral students. Research programs that will benefit from the cluster upgrade include development, application, and dissemination of computational tools in a diverse set of nine areas: (i) protein structure prediction and protein folding simulations, (ii) protein ligand discovery, (iii) integrative structure determination of macromolecular assemblies, (iv) structure determination by X-ray crystallography and cryo- electron microscopy, (v) bioinformatics and visualization of protein sequences, structures, and molecular networks, (vi) systems biology of biological networks, (vii) mammalian genetics, (viii) metagenomics, and (ix) bioimaging by magnetic resonance technologies. The cluster will make it possible to develop methods, carry out calculations, and serve the worldwide community in ways that are not feasible on our current system, due to limited speed and memory. The projects will provide (i) web-enabled tools and databases for biologists, (ii) insight into fundamental underpinnings of biochemistry and biology, as well as (iii) develop techniques with broad impact across multiple disciplines. In addition to the users, the cluster will benefit the many collaborators of the participating research groups, th students and faculty yet to join UCSF who will use the instrument, the participants in the UCSF high school and undergraduate diversity efforts, and the many worldwide users of our software, databases, and web services that we make freely available to the scientific community.
The proposed computer cluster will be used by at least 12 NIH-funded Principal Investigators of R01 grants and will support at least 83 active and pending NIH grants. Research programs span the fields of protein ligand discovery, structural biology, bioinformatics, systems biology, genomics, bioimaging, and medical informatics. The computer cluster will make it possible to carry out computationally intensive biomedical research not otherwise feasible.
Showing the most recent 10 out of 22 publications
