Big Data Repository for Structural Biology: Challenges and Opportunities by Piotr Sliz
- 1. Big Data Repository for Structural Biology: Challenges and Opportunities Piotr Sliz, PhD sliz@hkl.hms.harvard.edu ! SBGrid: http://sbgrid.org SBGrid Data Bank: http://data.sbgrid.org Twitter: @SBGrid YouTube: SBGridTV SBGrid Consortium Support Center at Harvard Medical School 300 Research Groups 13 Countries Long Term Sustainability: Membership Fee Harvard Medical! School
- 2. SBGrid supports compilation, installation and upgrades of ~300 scientific applications Several Software Categories (EM, NMR, Xrays, Comp Chem, etc.) Multiple versions of most applications OS X (10.6-10.10) and Linux support (CentOS 5-7) No additional, end-user configuration required Software always works = more time for research Core Mission: Grid Computing (Open Science Grid VO + Grid Portal) General Research Infrastructure (Boston Area) Training (workshops, software cataloguing, webtales) Webinars at youtube.com/SBGridTV Developer Resources Advocating for Open Source Software Morin et al. Shining Light into Black Boxes. Science, 2012. Other Activities: Additional! Publications Primary Citation: Other Citations:
- 3. New Opportunity: Data anonymous SBGrid member 1: “we cannot find the original frames for many of our structures (move from X to Y), including recent high impact projects. What do you recommend that we do?” anonymous SBGrid member 2: “I was able to locate the data directory but I must have done a good job cleaning up the disk space before I left: usually there are only two .img files left in the data directory, the 1st and the last image of a full run.” Lack of Storage Support for Diffraction Images derive reproduce improve correct • Stokes-Rees, I., Levesque, I., Murphy, F.V., Yang, W., Deacon, A., and Sliz, P. (2012). Adapting federated cyberinfrastructure for shared data collection facilities in structural biology. J Synchrotron Radiat 19, 462–467. • Terwilliger, T.C., and Bricogne, G. (2014). Continuous mutual improvement of macromolecular structure models in the PDB and of X-ray crystallographic software: the dual role of deposited experimental data. Acta Crystallogr. D Biol. Crystallogr. 70, 2533–2543. • Terwilliger, T.C. (2014). Archiving raw crystallographic data. Acta Crystallogr D Biol Crystallogr. • Guss, J.M., and McMahon (2014). How to make deposition of images a reality. Acta Crystallogr. D Biol. Crystallogr. 70, 2520–2532
- 4. Focus on Primary Data SBGrid Data Bank. Pilot: May 1st, Production: June 1st, 2015 EZID Dataset Lock BIODBCORE-‐000683 re3data.org Data Mining and Annotation
- 5. Web Interface Related! Datasets Depositors: URL: data.sbgrid.org Dataset Landing Page DataCite! Schema CC0 License Download Dataset URL
- 7. Data Access Alliance: Make Data easily accessible for reprocessing Minimize Project Cost Increase Redundancy Challenges Dataset Size (APIs, Data Access Alliance) Journal + Data Automation automated embargo release cross-referencing coordination/communication with journals Data vs Journal Citations Metrics: Dataset Deposition Rates Data Use: DAA Membership vs. direct downloads Dataset Quality (Level 0-2) Data Citations Master Format OME-TIFF vs DataCite vs DataVerse schema Transition to a Research Data Management Software ORCID integration and adoption
- 8. Opportunities Better support to ~300 structural biology laboratories: Compliance Reproducibility Integration with PDB and other repositories Other data types in addition to X-ray diffraction Thank you Piotr Sliz, PhD sliz@hkl.hms.harvard.edu ! SBGrid: http://sbgrid.org SBGrid Data Bank: http://data.sbgrid.org ! Twitter: @SBGrid YouTube: SBGridTV Stephanie Socias Pete Meyer Merce Crosas