ReproNim

ReproNim
Name	ReproNim

ReproNim ReproNim is a research initiative focused on reproducibility and neuroimaging methodology, integrating software, standards, and training to improve scientific practice. It collaborates with academic institutions, technology projects, and funding bodies to develop tools, workflows, and educational materials for neuroimaging researchers. The initiative intersects with efforts in open science, data standards, and computational reproducibility.

Overview

ReproNim operates at the intersection of initiatives like Open Science Framework, Brain Imaging Data Structure, Neuroimaging Informatics Technology Initiative, Human Connectome Project, and FMRIB Software Library, aligning with projects such as BIDS Apps, Nipype, AFNI, SPM, and FreeSurfer. It engages communities involved with National Institutes of Health, National Science Foundation, Wellcome Trust, European Research Council, and Horizon 2020, while interacting with platforms like GitHub, Zenodo, Figshare, Dataverse, and OSF Storage. Collaborations often reference repositories and standards exemplified by NeuroVault, OpenNeuro, XNAT, INDI, and COINS.

History

The development of ReproNim traces through collaborative movements similar to those behind Brainhack, Neurohackademy, OHBM, and INCF workshops, echoing influences from initiatives linked to Allen Institute for Brain Science, Human Brain Project, NIH BRAIN Initiative, and UK Biobank. Contributors have included researchers associated with institutions like Massachusetts General Hospital, Harvard Medical School, University of California, San Diego, University of Pennsylvania, and McGill University. Funding and policy context involves agencies such as NIH, NSF, European Commission, and foundations like Gordon and Betty Moore Foundation and Simons Foundation.

Projects and Tools

ReproNim develops and integrates tools comparable to Docker, Singularity, Conda, Python, R (programming language), and workflow engines akin to Nextflow and Snakemake. Tooling emphasizes compatibility with analysis suites like Freesurfer, AFNI, FSL, SPM12, and workflow frameworks such as Nipype and BIDS Apps. It promotes metadata capture practices akin to PROV, NIDM, DICOM, and standards propagated by BIDS, while interfacing with repositories including OpenNeuro, NeuroVault, Zenodo, GitLab, and Figshare.

Organizational Structure and Funding

Organizationally, the initiative mirrors consortia structures seen at INCF, OHBM, ISCB, AAAS, and university centers like Center for Reproducible Biomedical Modeling and networks similar to Data Science Institute at Columbia University. Funding models draw on grants from entities such as National Institutes of Health, National Science Foundation, Wellcome Trust, European Research Council, and philanthropic support from Gordon and Betty Moore Foundation and Simons Foundation. Governance often includes academic principal investigators from institutions like Johns Hopkins University, University of California, Berkeley, Yale University, University College London, and McGill University.

Adoption and Impact

Adoption of practices promoted by the initiative is observable in repositories such as OpenNeuro, publications in journals like NeuroImage, Nature Neuroscience, PLOS Computational Biology, Scientific Data, and in community events including OHBM annual meeting, Brainhack Global, Neurohackademy, and workshops at Society for Neuroscience meetings. Its influence relates to policy shifts at funders like National Institutes of Health and publishers including Nature, Science, PNAS, eLife, and PLOS One, contributing to reproducibility efforts similar to those driven by Center for Open Science and Reproducibility Project.

Criticism and Limitations

Critiques mirror debates in reproducibility movements engaging stakeholders such as Nature, Science, PNAS, eLife, and advocacy groups like Center for Open Science and Faculty of 1000. Practical limitations relate to computational infrastructures offered by Amazon Web Services, Google Cloud Platform, Microsoft Azure, and institutional clusters at universities like Stanford University, MIT, and University of Oxford, as well as to challenges discussed in fora like bioRxiv, arXiv, and policy venues such as National Academies of Sciences, Engineering, and Medicine reports. Concerns include sustainability reflected in funding patterns of agencies like NIH and NSF, scalability in contexts similar to Human Connectome Project data, and community uptake issues observed in collaborations with platforms like GitHub and Zenodo.

Category:Neuroimaging