Open Neuro

Open Neuro
Name	Open Neuro
Formation	2016
Type	Repository
Headquarters	Santa Clara County, California
Region	International

Open Neuro is an open data repository for neuroimaging and neuroscience datasets designed to promote transparency, reproducibility, and data sharing across the neuroscience community. It facilitates the archiving and dissemination of functional magnetic resonance imaging datasets, structural MRI datasets, magnetoencephalography datasets, and related behavioral and metadata, enabling reuse by researchers, consortia, and institutions worldwide. The platform integrates with community standards, computational services, and scholarly outlets to connect data producers with analysts in academia, industry, and public research initiatives.

Overview

Open Neuro functions as a centralized archive supporting dataset deposition, curation, versioning, and public access. It interoperates with tools and initiatives such as the Brain Imaging Data Structure, Neuroimaging Informatics Tools and Resources Clearinghouse, International Neuroinformatics Coordinating Facility, OpenfMRI, and DataCite to ensure persistent identifiers and metadata quality. The repository is used by investigators affiliated with institutions like Harvard University, Stanford University, Massachusetts Institute of Technology, University College London, and University of Oxford and supports collaborations with consortia including the Human Connectome Project, Allen Institute for Brain Science, ENIGMA Consortium, ADNI, and BRAIN Initiative. It aligns with publishing practices of journals such as Nature Neuroscience, Neuron (journal), Scientific Data (journal), PLOS ONE, and eLife (journal).

History and Development

The project emerged from community efforts to improve reproducibility in neuroimaging following landmark events and reports such as work by John Ioannidis, discussions at meetings organized by Organization for Human Brain Mapping, and data-sharing advocacy by groups including Open Science Framework and Center for Open Science. Early technical foundations built on precedents like OpenfMRI and infrastructure from Amazon Web Services grants, with institutional contributions from University of California, Berkeley, University of California, San Diego, and Stanford University. Governance and funding have intersected with agencies and programs such as the National Institutes of Health, National Science Foundation, Wellcome Trust, Alfred P. Sloan Foundation, and philanthropic partners like Gordon and Betty Moore Foundation. The platform’s roadmap has been influenced by standards discussions at venues like Neurohackweek, conferences such as Society for Neuroscience, and workshops hosted by Global Brain Consortium.

Data Repository and Standards

Datasets are organized and validated against the Brain Imaging Data Structure standard and annotated with controlled vocabularies drawing on resources like NeuroLex, Resource Description Framework, Dublin Core, and identifiers from ORCID and DataCite. Metadata integration leverages ontologies developed by groups including Neuroscience Information Framework, Gene Ontology Consortium, and Human Brain Project working groups. The repository supports data types aligned with protocols from manufacturers and initiatives such as Siemens Healthineers, GE Healthcare, Philips Healthcare, Magstim, and acquisition projects like OpenNeuro-MEG and datasets contributed by centers including Massachusetts General Hospital and Johns Hopkins University. Persistent identifiers and citation practices follow guidelines from CrossRef and data citation principles endorsed by Force11.

Platform Features and Tools

Open Neuro provides dataset search, command-line clients, web upload interfaces, and integration with analysis platforms and workflow engines including BIDS Apps, fMRIPrep, AFNI, FSL, FreeSurfer, SPM (software), Nipype, and Docker (software). Computational linkage enables use with cloud compute services like Google Cloud Platform, Amazon Web Services, and research computing environments such as XSEDE. The platform facilitates reproducible pipelines using tools including GitHub, GitLab, Zenodo, Jupyter Notebook, Neurostars, and Binder (service), and supports provenance tracking compatible with PROV (W3C). Data accessibility is enhanced through APIs and connectors used by platforms like OpenNeuroDerivatives, CBRAIN, and BrainLife, and interoperability with analytic frameworks such as Python (programming language), MATLAB, R (programming language), Julia (programming language), and machine learning libraries including TensorFlow, PyTorch, and scikit-learn.

Governance, Ethics, and Privacy

Governance and stewardship involve stakeholders drawn from universities, research institutes, and non-profit organizations such as International Neuroinformatics Coordinating Facility affiliates, with community governance models influenced by policies from National Institutes of Health data-sharing requirements, ethical frameworks debated at Council for International Organizations of Medical Sciences, and consent standards exemplified by templates from Open Humans and Informed Consent. Privacy-preserving practices incorporate de-identification workflows, defacing algorithms informed by research from FreeSurfer developers and groups at University of Pennsylvania, and compliance considerations related to laws and regulations including Health Insurance Portability and Accountability Act and international norms discussed at bodies like the European Data Protection Supervisor and initiatives such as Global Alliance for Genomics and Health. Data use agreements and licensing strategies reflect recommendations from organizations like Creative Commons and researchers engaged in debates highlighted in publications by Nature Communications and Science (journal).

Impact and Community Adoption

The repository has been cited in studies across cognitive neuroscience, clinical neuroimaging, computational psychiatry, and connectomics by researchers at institutions including University of California, Los Angeles, Yale University, Columbia University, University of Toronto, McGill University, and Karolinska Institutet. It supports reproducible publications tied to datasets referenced in proceedings of Human Brain Mapping, International Conference on Cognitive Neuroscience, and workshops at NeurIPS and OHBM. Community engagement includes contributions from research groups such as WUSTL, UCSF, McLean Hospital, Broad Institute, and collaborations with industry partners like Google Research and IBM Research. Adoption metrics are discussed in analyses published in venues such as Nature Methods, PLOS Computational Biology, and Frontiers in Neuroscience and inform training programs at Neurohackademy and summer schools including INCF Neuroinformatics Training.

Category:Neuroscience data repositories