Issues in Biomedical Research Data Management and Analysis: Needs and Barriers

Anderson, Nick; Lee, E. Sally; Brockenbrough, J. Scott; Minie, Mark; Fuller, Sherrilynne S.; Brinkley, James F.; Tarczy‐Hornoch, Peter

doi:10.1197/jamia.m2114

Cited by 135 publications

(118 citation statements)

References 24 publications

Supporting

Mentioning

113

Contrasting

Unclassified

Order By: Relevance

“…Cragin et al (2010) also found that researcher attitudes towards research data sharing were colored by the amount of time it would take to prepare data for sharing. Anderson et al (2007) report that scientific communities are interested in solving these problems themselves, but lack the tools and resources to do so. This duality between skepticism of the regulatory framework for research data management and a strong sense of urgency to safeguard data is a key point in understanding how to approach research data on campus.…”

Section: Interview Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Assessing Research Data Management Practices of Faculty at Carnegie Mellon University

Tuyl

Michalek

2015

Journal of Librarianship and Scholarly Communication

View full text Add to dashboard Cite

INTRODUCTION Recent changes to requirements for research data management by federal granting agencies and by other funding institutions have resulted in the emergence of institutional support for these requirements. At CMU, we sought to formalize assessment of research data management practices of researchers at the institution by launching a faculty survey and conducting a number of interviews with researchers. METHODS We submitted a survey on research data management practices to a sample of faculty including questions about data production, documentation, management, and sharing practices. The survey was coupled with in-depth interviews with a subset of faculty. We also make estimates of the amount of research data produced by faculty. RESULTS Survey and interview results suggest moderate level of awareness of the regulatory environment around research data management. Results also present a clear picture of the types and quantities of data being produced at CMU and how these differ among research domains. Researchers identified a number of services that they would find valuable including assistance with data management planning and backup/storage services. We attempt to estimate the amount of data produced and shared by researchers at CMU. DISCUSSION Results suggest that researchers may need and are amenable to assistance with research data management. Our estimates of the amount of data produced and shared have implications for decisions about data storage and preservation. CONCLUSION Our survey and interview results have offered significant guidance for building a suite of services for our institution.

show abstract

Section: Interview Resultsmentioning

confidence: 99%

“…A third option, taken by some in the literature, is to combine surveys and interviews to leverage the respective breadth and depth provided by these methods (e.g. Anderson et al, 2007;Beagrie et al, 2009). …”

Section: Literature Reviewmentioning

confidence: 99%

Assessing Research Data Management Practices of Faculty at Carnegie Mellon University

Tuyl

Michalek

2015

Journal of Librarianship and Scholarly Communication

View full text Add to dashboard Cite

show abstract

“…Some of the problems have been well documented by Anderson et al [3] in the context of biomedical research. Many of their conclusions apply in general across the biological sciences, especially at the medium sized laboratory level.…”

Section: The Data Handling Problemmentioning

confidence: 97%

DataPipeline: Automated importing and fitting of large amounts of biophysical data

Farrell

2012

J Comput Chem

View full text Add to dashboard Cite

Raw data from experiments across the biological sciences comes in a large variety of text formats. In small or medium sized laboratories researchers often use an assorted collection of software to interpret, fit, and visualize their data. The spreadsheet is commonly the core component of such a workflow. The limitations of such programs for large amounts of heterogeneous data can be frustrating. We report the construction of DataPipeline, a desktop and command‐line application that automates the tasks of importing, fitting, and plotting of text‐based data. The software is designed to simplify the process of importing text data from various sources using simple configuration files to describe raw file formats. Once imported, curve fitting can be performed using custom fitting models designed by the user inside the application. Fitted parameters can be grouped together as new datasets to be fitted to other models and experimental uncertainties propagated to give error estimates. This software will be useful for processing of data from high through‐put biological experiments or for rapid visualization of pilot data without the need for a chain of different programs to carry out each step. DataPipeline and source code is available under an open source license. The software can be freely downloaded at http://code.google.com/p/peat/downloads/list. © 2012 Wiley Periodicals, Inc.

show abstract

“…It is inevitable for the future biomedical domain expert to switch from the classical consumer-like role [44] to an active part in the knowledge discovery process [27,30]. However, this is not so easy, because it is well known that many biomedical research projects fail due to the technical barriers that arise to the domain experts in data integration, data handling, data processing, data visualization and analysis [1,34,43]. A survey from 2012 among hospitals from Germany, Switzerland, South Africa, Lithuania, and Albania [60] showed that only 29 % of the medical professionals were familiar with any practical application of data mining methods and tools.…”

Section: Graph-based Discovery In Medical Researchmentioning

confidence: 99%

Knowledge Discovery from Complex High Dimensional Data

Lee

Holzinger

2016

Solving Large Scale Learning Tasks. Challenges and Algorithms

View full text Add to dashboard Cite

Abstract. Modern data analysis is confronted by increasing dimensionality of problems, mainly contributed by higher resolutions available for data acquisition and by our use of larger models with more degrees of freedom to investigate complex systems deeper. High dimensionality constitutes one aspect of "big data", which brings us not only computational but also statistical and perceptional challenges. Most data analysis problems are solved using techniques of optimization, where large-scale optimization requires faster algorithms and implementations. Computed solutions must be evaluated for statistical quality, since otherwise false discoveries can be made. Recent papers suggest to control and modify algorithms themselves for better statistical properties. Finally, human perception puts an inherent limit on our understanding to three dimensional spaces, making it almost impossible to grasp complex phenomena. For aid, we use dimensionality reduction or other techniques, but these usually do not capture relations between interesting objects. Here graph-based knowledge representation has lots of potential, for instance to create perceivable and interactive representations and to perform new types of analysis based on graph theory and network topology. In this article, we show glimpses of new developments in these aspects.

show abstract

Issues in Biomedical Research Data Management and Analysis: Needs and Barriers

Cited by 135 publications

References 24 publications

Assessing Research Data Management Practices of Faculty at Carnegie Mellon University

Assessing Research Data Management Practices of Faculty at Carnegie Mellon University

DataPipeline: Automated importing and fitting of large amounts of biophysical data

Knowledge Discovery from Complex High Dimensional Data

Contact Info

Product

Resources

About