Computational Resources
IGSP offers researchers a broad set of computational and data management resources for genome research. A core information technology staff administers the server infrastructure and develops resources that complement the biological interests of researchers within the institute. The computational infrastructure is being developed to support integrated systems and sharing of resources among the wide range of investigators in IGSP.
Computing by the numbers. Currently the computational systems feature closely integrated Linux systems running on 32-bit and 64-bit architectures with access to more than 12 terabytes of disk space, most of it in RAID. Workstations throughout the physical facilities of IGSP run Windows, Mac OS X, and Linux operating systems, with the primary platform for computational biology being Linux. Four full-time staff maintain servers and workstations.
Computer integration. The IGSP computing infrastructure is designed to make research collaborations easier. Although the process of sharing data between computing platforms is a challenge, the system that is in place and that is being further developed makes collaborative work possible between labs and individuals using completely different computer platforms.
Since genome sciences and policy are by nature interdisciplinary, we seek to remove technological barriers that impede interdisciplinary work.
Applications. Standard computational biology/bioinformatics software is generally available, and over the past five years, researchers have used systems to do genome assembly and annotation, gene expression analysis, comparative genomics, phylogenetics, creation of high-throughput data creation and analysis pipelines, and the like. IGSP participates in the development of the cluster computational resource spearheaded by the Duke's Computational Science, Engineering and Medicine (CSEM) group.