Science & Innovation

Summer of Sharing


summer of sharing

Open science initiatives are transforming the neuroimaging community. Researchers who once struggled to obtain 20-30 datasets now have unrestricted access to thousands of scans, including data obtained from developing, aging, and clinical populations. Equally important, the sharing of data analysis scripts and code is becoming increasingly common, thereby enhancing the dissemination of knowledge and capabilities among laboratories—as well as facilitating replication efforts. Although exciting, we are only at the beginning. The success of open science initiatives remains dependent upon continued community participation.

Against this background, the International Neuroimaging Data-sharing Initiative (INDI) is pleased to announce the launch of its 2012 Summer of Sharing—an effort dedicated to the sustenance and acceleration of data and analytic resource sharing among imaging community members. In order to accomplish its goal the INDI Summer of Sharing initiative is requesting pledges for the contribution of:

  1. Resting state fMRI data employed in published studies, along with available meta-data.
  2. Diffusion imaging data employed in published studies, along with available meta-data.
  3. Analysis scripts employed in published studies, as well as those currently employed by laboratories.

We are actively asking investigators around the world to get involved in the sharing process. Already, the following resources are scheduled for sharing this summer:

  • Adelstein et al., 2011 (PLoS One): R-fMRI, anatomical and NEO-V Personality Inventory phenotypic data, from the recently their published manuscript.

  • ART (Artifact Rejection Toolbox). An open source program which facilitates detection and correction of artifacts for fMRI
    task and connectivity analyses, provides diagnostic tools which assist in appropriate design specification and is interoperable with fMRI analysis software packages (e.g., SPM, FSL, Conn).

  • The Autism Brain Imaging Data Exchange (ABIDE): a global, multisite consortium of laboratories dedicated to the study of autism that will release an aggregate dataset that is expected to consist of nearly 500 individuals with autism, and their matched typically developing controls.

  • Brain Genomics Superstruct Project: Anatomical, functional, behavioral and personality measures from 1500 young adult subjects will be released during the summer of 2012. All data were acquired on matched MRI platforms.

  • Brainhack: a Neuro-Bureau resource that points to collaborative projects in the field; the resource will be complemented by the Brainhack 2012 Unconference—a workshop that will blend the “unconference” and “hackathon” concepts to create a unique collaborative experience for participants interested in advancing analytic techniques and resources for functional and structural imaging.

  • Center for Biomedical Research Excellence (COBRE): examines the neural mechanisms of schizophrenia by integrating multiple neuroimaging methods with psychiatric, neuropsychological and genetic testing. COBRE is composed of four tightly integrated projects that will break new ground in schizophrenia research by combining neuroimaging data in a sophisticated and innovative way. COBRE is contributing anatomical and functional MR data from 72 patients and 75 healthy controls (ages ranging from 18 to 65 in each group) to the INDI, this summer.

  • Child Mind Institute Librarian Initiative: Comprehensive hand-vetted and sorted reference libraries for various literatures, including Resting State fMRI and Diffusion Tensor Imaging are now available via Mendeley, with monthly updates scheduled.

  • Collaborative Informatics Neuroimaging Suite (COINS): a database that offers tools to share existing data with other researchers, acquire data from other researchers, and manage studies from beginning to end. COINS currently manages over 400 studies with more than 232,000 assessments and 23,000 scans collected for 16,000 participants at the Mind Research Network, the Nathan Kline Institute, University of Colorado–Boulder, the Olin Neuropsychiatry Research Center, and other institutions.

  • Configurable Pipeline for the Analysis of Connectomes (C-PAC): Building on the success of the release of the 1000 Functional Connectomes Project analysis scripts, the INDI team will be releasing a plug-and-play Nipype-based pipeline package that is easily configurable to accomplish a broad array of resting-state fMRI analyses.

  • Conn V.13: fMRI Functional Connectivity Toolbox. Matlab toolbox for performing functional connectivity analyses of fMRI data. Features: ROI-to-ROI, seed-to-voxel, voxel-to-voxel (whole brain connectome), graph-theory approaches to resting state and task-related connectivity analyses, using both GUI and batch processing. Spatial preprocessing and physiological correction available (e.g., CompCor), as well as stastistical analysis (e.g., first-level univariate and multivariate regression, bivariate and semipartial correlation connectivity measures, random effects and mixed
    effects models).

  • Connectome Computation System (CSS): It provides a common platform for brain connectome analysis by integrating the functionality of AFNI, FSL, Freesurfer, extending the utility of FCP scripts and by integrating the brain surface information reconstructed by Freesurfer. It thus can translate all connectome (particularly the functional connectome) analyses onto a manifold of the cortical surface.

  • DPARSF and REST: With the new releases of MATLAB-based user-friendly pipelines of Data Processing Assistant for Resting-State fMRI (DPARSF 2.2) and Resting-State fMRI Data Analysis Toolkit (REST 1.8), users can acquire resting-state fMRI measures efficiently (by parallel computing) and conveniently (by GUI or command line).

  • Eklund et al., 2012 (Neuroimage): Datasets, Matlab scripts and results from their analysis of the appropriateness of parametric approaches for the modeling of temporal correlations in fMRI timeseries data, are being shared at

  • The Enhanced NKI Rockland Sample: a recently launched large-scale, R-fMRI and DTI cross-sectional, community ascertained sample of individuals between 6 and 85 years of age, accompanied by a deep phenotypic protocol that broadly samples psychiatrically relevant behavioral domains.

  • Jones DT et al. (2012) (PLoS One): The authors are making available a high and low-dimensional ICA data-set obtained from a population-based sampling of 892 cognitively normal older subjects participating in the Mayo Clinic Study of Aging. These aggregate components can be used for spatial-temporal dual regression or the corresponding ROIs can be used as a functional atlas. Files can be found at:

  • LORIS: A modular and extensible web-based data management system that integrates all aspects of a multi-center study: from heterogeneous data acquisition (imaging, clinical, behaviour, and genetics) to storage, processing, and ultimately dissemination. Initially developed to manage data for the NIH MRI Study of Normal Brain Development (Evans and Brain Development Cooperative Group, 2006,, LORIS has since been adapted and implemented in numerous decentralized large-scale studies internationally.

  • Mindboggle Data and Software: Python software designed to automate shape analysis and labeling of human brain anatomy in MR image data.

  • NeuroImaging Datasharing Hall of Fame: Have you ever wondered how much resources have been saved by reusing already acquired data? Do you struggle with convincing others about the benefits ofdatasharing? This project aims ad crowd sourcing information about studies that were based on shared data and attempts to estimate how much would they would cost otherwise. Source code is available at

  • Nipype Connectivity Workshop: A two-day workshop focusing on Nipype—an open framework that helps in the designing, maintaining, sharing and executing data processing workflows. The workshop will take place in Magdeburg, Germany on September 8-9, 2012.

  • NITRC-IR: An XNAT-based Image Registry hosted at NITRC to support image sharing for NITRC projects. Offers a cloud-based federated data storage system for sharing neuroimaging data in DlCOM and NIFTI formats. Users can search for stored images based on demographics, imaging parameters, and fMRI descriptors. From search results, users can select and download images. NITRC-IR is an implementation of XNAT. 1000 Functional Connectomes Classic, ADHD-200 and CANDI Share: Schizophrenia Bulletin 2008 datasets are currently available in the NITRC-IR. In the future, other NITRC tools/resources will be able to make their data sets available to NITRC-IR.

  • OpenfMRI project: A data-sharing repository for task-based fMRI datasets.

  • PANDA: A matlab pipeline toobox for analyzing brain diffusion images. It provides fully-automatic processing from raw diffusion-MRI dataset (DICOM/NIFTI) of any number of subjects to final outputs,  which include 1) diffusion metric data (e.g. FA) ready for statistical analysis; 2) brain white matter networks based on deterministic or probabilistic tractography. Particularly, PANDA supports both sequential and parallel computing modes, allowing for maximum usage of available computing resource. The parallel environment can be a single desktop with multiple-cores or a computing cluster with a SGE system. In addition, PANDA has a very friendly GUI.

  • Power et al. 2011 (Neuron): The authors are now sharing the summary community assignments from their work on fcMRI networks, and will release a suite of scripts through NITRC this summer, so that others can replicate these analyses and carry out related work. Additionally, later this summer, the dataset used for this work will be released through INDI.

  • Shirer et al. (2011, Cerebral Cortex): Resting-state and subject-driven task data are being provided for 26 subjects, and structural MRI data for 14 of those 26. Additionally, the functional data will be distributed in two forms: physio-regressed and physio-not-regressed. Coming soon.

  • Tissue Stack: An open source web based image viewer which allows the user to view 3D data in the traditional triplanar views. While at its core it's modeled after GIS web mapping applications, it's intended use is for neuroimaging. Tissue Stack aims at serving its data as image tiles which can be both pre-tiled in advance or created on the fly by extracting from the original source file in either MINC or NifTI-1 format. A blog for Tissue Stack is based at

  • Yeo et al. 2011 (J. Neurophysiol) andBuckner et al. 2011 (J. Neurophysiol): The authors have made available their cortical and cerebellar parcellations in multiple data formats

It is our hope that this list is just the beginning. We will follow with updates throughout the summer months. The success of open science depends on all of us—so get involved, and Share That Brain!

Contact us at to make a pledge or learn more about how to get involved.

Please help us improve the Symptom Checker!

Click here to share your thoughts about using the tool.