ReproNim: A Center for Reproducible Neuroimaging Computation to support Resource Discovery, Interoperability, and Replicable Results
-
1
TCG, Inc., United States
-
2
MIT, United States
-
3
UC San Diego, United States
-
4
Dartmouth College, United States
-
5
University of Massachusetts Medical School, Psychiatry Neuroinformatics, United States
-
6
UC Irvine, United States
-
7
UC Berkeley, United States
Introduction
Over the last two decades a vast technological, computational, and societal infrastructure has emerged and transformed how information is collected and knowledge is gathered in all facets of science. In the scientific process, methods should be reproducible. In the economics of science, data and methods should be maximally reusable in order to maximize the scientific return on the data acquisition investment and results should be reliable in order to build solutions and develop new research directions.
Approach
Our approach is divided into three technology development efforts and a training mission. 1) Resource Discovery: In this project, we will develop new search and discovery tools to create a sophisticated, comprehensive and dynamic search environment for working with distributed neuroimaging data, tools, workflows and execution environments. These tools will integrate and extend existing successful community platforms, data, and search services. This work will specifically support the end user with a specific analytic goal to find the appropriate data and analysis workflows that can subsequently executed on local or cloud-based executions environments. 2) Data Modeling and Interoperability: The primary aims of this project will be to provide a consistent and extensible data model for communicating information in brain imaging and associated easy to use, supported, and documented software libraries and services for developers. Using these models, we will provide end users a set of commonly used reproducible workflows with integrated provenance tracking that are easy to use. The workflows will be validated and ideally be executable on infrastructure available to researchers. The workflows will generate queryable results using standardized data models that are essential to allow software and people to communicate and interpret data precisely. 3) Reusable Execution Environments: The goal of this project is to enable reproducible computation through full automation and tracking of computing environments. The result will be a NeuroImaging Computation Environments MANager (NICEMAN). NICEMAN will support easy and reproducible execution of neuroimaging analysis workflows on various computational platforms, while efficiently reusing and integrating existing free and open-source software products and data sharing initiatives. Execution of workflows will be achieved by automatically creating computation environments where necessary and making sure software and datasets are available, executing the workflow(s), and providing results and detailed provenance information about the environment back. 4) Training: Our objectives are to provide the brain imaging community with online training materials based on the concepts and software developed by the center. We will conduct training workshops to teach the fundamentals of reproducible neuroimaging and to use center resources and tools effectively. In particular, we aim at providing researchers and clinicians with the know-how to integrate a full cycle of research: hypothesis testing, data and software discovery, finding and adapting “playbooks” or pipelines, allowing provenance tracking and results discovery, and reproducible computations. This training will cultivate a clear understanding of the concepts, assumptions, and limitations underlying the reproducible research automation tools.
Conclusion
Together, these technologies support a vision of a neuroimaging research landscape where the generation of knowledge is accomplished in a reproducible fashion (in terms of data, analysis, and computation) coupled with the ability to discover, reuse, and extend these studies by others in the community.
Acknowledgements
Funding for this project was provided by NIH grant: NIBIB P41 EB019936 “Center for Reproducible Neuroimaging Computation - CRNC” PI: Kennedy.
Keywords:
Neuroimaging,
Reproducibility of Results,
computational modeling,
data sharing,
data modeling
Conference:
Neuroinformatics 2016, Reading, United Kingdom, 3 Sep - 4 Sep, 2016.
Presentation Type:
Poster
Topic:
Infrastructural and portal services
Citation:
Buccigrossi
R,
Ghosh
SS,
Grethe
JS,
Halchenko
YO,
Haselgrove
C,
Keator
DB,
Kennedy
DN,
Martone
M,
Poline
J,
Preuss
N and
Travers
M
(2016). ReproNim: A Center for Reproducible Neuroimaging Computation to support Resource Discovery, Interoperability, and Replicable Results.
Front. Neuroinform.
Conference Abstract:
Neuroinformatics 2016.
doi: 10.3389/conf.fninf.2016.20.00083
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
31 May 2016;
Published Online:
18 Jul 2016.
*
Correspondence:
Dr. David N Kennedy, University of Massachusetts Medical School, Psychiatry Neuroinformatics, Worcester, MA, 01605, United States, David.Kennedy@umassmed.edu