The Georgia Institute of Technology Data Curation Program

1. The Georgia Institute of Technology Data Curation Program Tyler O. Walters, Associate Director, Technology & Resource Services Library & Information Center, Georgia Institute of Technology For NSF Site Visit to MIT, February 8 2010

2. Why is GT involved in DataSpace Project? To establish a system of federated, distributed data repositories. (GT: recognizes need for this research infrastructure) Shared background with MIT, DSpace development since 2003 Collaborate on data curation research and services, start with neurosciences and biosciences GT is committed to advancing its data curation program (created “research data librarian” position)

3. Neuroscience and Neuroimaging at GT GT/GSU Center for Advanced Brain Imaging (CABI) http://www.cabiatl.com/CABI/ (Chris Rordan, Director) • CABI = 27 PIs / faculty + 35-40 researchers, Center holds ca. 120 TB • Neuroscience: leading example of a domain that will curate its data in a diffuse fashion; hence, university-level solutions will become significant • fMRI brain studies -- Formats: DICOM, NIfTI, and EEG numeric data • Work with Prof. Paul Corballis: http://psychology.gatech.edu/corballislab/ • Data Dissemination: publisher rules vary -- researchers desire linking e-publishing activities with final data, however, they struggle with how best to enact primary-secondary source relationship

4. GT Library:Engagement in Data Curation • 2007-09: Library Data Curation Work Group • Interviewed researchers about data practices/needs, collected interview data • 2009-10: Research Data Project Librarian & Workgroup • Continue assessments of faculty data practices (using DAF) • Library’s Digital Development Team • Assessing & implementing technology infrastructure for data curation • Core systems for data curation • Sun StorageTek 2540 disk arrays / SL 500 Tape Library / Sun SAM server & ZFS • Work with MIT technology stack for data curation • Extended storage and preservation services (external partners, e.g. MetaArchive, Chronopolis, consider new DuraCloud service)

5. Neuroimaging and GT Project Team MIT: Martinos Imaging Center / GT: Ctr. for Advanced Brain Imaging Synergies in data curation to advance science through data sharing, publishing, and preservation The GT Team: • Library: data curator, storage/network manager, programmer, repository librarian, psychology librarian, AD for technology (Walters) • OIT: director of infrastructure and architecture (Chen) • CABI: Prof. Corballis, graduate student • Advisors: Prof. David Bader, Exec. Director, High-Performance Computing Dr. Bill Underwood (GTRI), digital archives research Prof. Leo Mark (Computing), atmospheric science data curation

6. Data Curation Strategy • Data deposition/acquisition/ingest • SIPs prepared by CABI graduate student / GT Research Data Librarian • Data curation and metadata management • Collaborate on metadata guidelines, policies on access, retention, formats , etc. • Data protection (policies, tools, procedures) • Chen (OIT), Baines (OIT Info. Security), Helms and Walters (Library), Corballis (CABI) • Data discovery, access, use, dissemination • Collaborate on portal design, descriptive metadata for expert and citizen use • Data interoperability, standards, integration • Identify, develop, and use in-common ontologies, semantic frameworks, data transfer and integration protocols between partners • Data evaluation, analysis, and visualization • Build technical framework to incorporate researcher’s tools