Data reuse and computational reproducibility have become increasingly important in scientific work over the past decade. This webinar consists of two talks, each providing a different perspective on computational reproducibility. The first talk will focus on advancing and assessing the reproducibility of data and code deposited on a Dataverse data repository. The speaker will present a large-scale study of rerunning preserved code. Following the study outcomes, the speaker will present lessons and how to alleviate the observed lack of reproducibility. Also, she will give a refined approach for research dissemination for painless reuse, and ideas on how to signal quality in shared research. The second talk will share challenges and opportunities associated with research data and software management in support of the high performance computing-driven computational reproducibility of research. The speaker will then summarize lessons learned and good practices for facilitating reproducible research.
“Ana Trisovic is a Sloan postdoctoral fellow at Institute for Quantitative Social Science (IQSS). Her research focus is on computational reproducibility, data provenance and open science. She investigates how data repositories can facilitate reproducibility and what are the most common causes of research irreproducibility in deposited research. In these projects, she collaborates with Mercè Crosas and the Dataverse team. Previously, she was a postdoctoral fellow at the University of Chicago, where she worked with the Energy Policy Institute (EPIC) and the Library. She completed her PhD at the University of Cambridge in 2018 on the topic of “Data preservation and reproducibility at the LHCb experiment at CERN”. While at CERN, she worked with the LHCb collaboration, CERN Open Data and CERN Analysis Preservation groups. During her PhD, she was a scholar of the Muir Wood studentship of the Newnham College, CERN doctoral student program and Google Anita Borg Memorial Scholarship.
Qian Zhang is the 2018 CLIR postdoc fellow in software curation at the University of Waterloo, co-hosted by the University Library and the David R. Cheriton School of Computer Science. Her responsibilities include leading research into both data and software curation. In specific, Qian actively builds collaborative working relationships within the Library, and with faculty and research teams at the UW to establish a framework and a set of best practices for curating data and software code that requires preservation over time, primarily for their value in facilitating research and the reproducibility of research. Before joining the University of Waterloo, she worked as a postdoc across the RDS (Research Data Service) of the Library, iSchool and NCSA (National Center for Supercomputing Applications) at the University of Illinois at Urbana-Champaign.”
This webinar will be presented in English and recorded. Engagement during the webinar in both official languages is welcomed. Previous webinar recordings are available on the CARL YouTube Channel. Links to the recordings and slides can also be found on Portage Training Resources.