Brain-CODE is a large-scale informatics platform that manages the acquisition and storage of multidimensional data collected from participants with a variety of brain disorders.
Ontario’s Big Data Opportunity
The Ontario Brain Institute is a provincially‐funded, not‐for‐profit research centre seeking to maximize the impact of neuroscience and establish Ontario as a world leader in brain research, commercialization and care. Convergent partnerships are created between researchers, clinicians, industry, patients, and their advocates to foster discovery and deliver innovative products and services that improve the lives of those living with brain disorders. To help realize this goal, the OBI funds five Integrated Discovery Programs (ID Programs) spanning the areas of epilepsy, cerebral palsy, neurodegenerative disorders, depression and neurodevelopmental disorders. The creation of these programs has resulted in a big data opportunity.
big da·ta (noun): extremely large datasets that may be organized computationally to reveal patterns, trends, and associations, especially relating to human behaviour and interactions.
The creation of these programs has resulted in a Big Data opportunity. Integrating and sharing the wealth of Ontario-based brain research data has immense potential to inspire innovative, impactful diagnostics and treatments for brain disorders. To facilitate collaboration and discovery, OBI has developed Brain-CODE. Brain-CODE is an extensible informatics platform that manages the acquisition, storage and sharing of multidimensional data collected from patients with a variety of brain disorders (Figure 1). As an ID program researcher, Brain-CODE is your data storage solution, analytics toolbox, and collaboration hub. Your commitment to data sharing will transform Brain-CODE into a platform for discovery.
By collecting data elements across diseases in a standardized manner this complex matrix arises: not only can deep comparisons across data modalities be made within one single brain disorder, but so too can deep comparisons begin to be investigated across brain disorders. Brain-CODE aims to facilitate data-driven discovery in neuroscience unlike ever before.
Brain-CODE's Guiding Principles
1) Privacy & Security
Brain-CODE adheres to high levels of data privacy and security. Encryption and de-identification tools are used to protect participant data. To federate with other databases, Brain-CODE has a high-security data transfer infrastructure using enhanced validation certificates at the Centre for Advanced Computing (CAC) facility which meets Code of Federal Regulations (CFR) Title 21, Part11 standards.
2) Open & Collaborative
Brain-CODE was designed to enable data sharing and to foster collaborations among a broader researcher community. Allowing researchers to make their data open to other researchers studying the same or different disorders is becoming a common approach that leads to ground-breaking discoveries.
3) Interdisciplinary & Integrated
Brain-CODE facilitates data sharing within and across brain disorders to further understand mechanisms of disease, commonalities, comorbidities, while promoting novel hypothesis generation. This integrated approach to research sparks collaboration between clinicians, researchers, basic scientists, industry partners, participants and patient advocacy groups.
4) Patient Centred
The fundamental purpose of data sharing and collaboration on Brain-CODE is to enable greater insights into the data and ultimately drive faster translation of the discoveries towards positive patient outcomes.
Standardization is a high priority for Brain-CODE and it ensures interoperability of datasets. Standardization efforts including clinical common data elements (CDEs), MEG and EEG standardization forms, and fBIRN phantom scans, help to define and format the vast array of clinical, neuroimaging and molecular data to optimize downstream integrative analyses.
6) Discovery for Health Impact
By facilitating collaboration on a platform of standardized neuroscience data, the stage is set for integrative big data analytics. Integration of datasets across disciplines has the potential to catalyze new discoveries which can be translated into new diagnostics and treatments for brain disorders.
At the Heart of Brain-CODE
Brain-CODE is developed by experts in research informatics
The Indoc Consortium is led by Indoc Research, who helps design and implement the technical infrastructure, working very closely with OBI on the development of the platform. Other members are the Electronic Health Information Laboratory (EHIL) led by Dr. Khaled El-Emam, a world-leader in data privacy and security, the Centre for Advanced Computing (CAC), which hosts Brain-CODE, and the Rotman Research Institute (RRI), that provide the neuroimaging informatics infrastructure and workflows for Brain-CODE.
Brain-CODE is currently hosted on high-performance computing nodes at the CAC which boasts a combined processing performance of 5 Tflops. The Brain-CODE network is connected through the Orion fibre-optic network, offering a transfer bandwidth of up to 10Gbps. As the use and requirements for Brain-CODE grow, additional hardware nodes can be allocated for increased data storage, specialized data processing, added demand for federation, and intensive concurrent analytical tasks.
The Brain-CODE neuroinformatics platform is designed around five key components: data capture, data storage, data integration, analytics, and federation. Electronic data capture software enables researchers to enter and store multi-modal data on one convenient web-based interface. REDCap and OpenClinica for clinical data, LabKey for molecular and genomics data, and SPReD (based upon XNAT) for imaging data, have been integrated within Brain-CODE to suit diverse research needs.
Integration of the diverse data sets is facilitated by a middleware application and standardized data infrastructure. IBM Infosphere enables the heterogeneous sets of research data to be regrouped and organized for analytics.
Finally, specialized software is also being developed for Brain-CODE to enable secure linking and federation with other national and international health databases.
The Brain-CODE team is working towards building a robust analytics capacity for researchers to analyze their data as well as share, collaborate, and analyze other data sets effectively and efficiently.
Customizable analytics workspaces with access to high performance computing capabilities are available to registered Brain-CODE users by request. These virtual workspaces ensure data is secure, meanwhile providing access to a host of analytics tools, including R and Python. Researchers may also bring their own software licenses into the workspace environment if additional analytics capacity is required.
Federation with National and International Databases
Brain-CODE is designed to seamlessly link with autonomous databases to leverage existing resources, augment data holdings and provide for richer analytics. This enables researchers to link their data with data that was previously difficult to access, making new discoveries possible.
- ICES - Brain-CODE Linkage:
- OBI and the Institute for Clinical and Evaluative Sciences (ICES) are currently working towards piloting a secure linkage which would allow researchers to augment the data collected from research participants with administrative health data.
- The linkage of data from ICES with an external database such as Brain-CODE is unprecedented.
- LORIS-Brain-CODE Federation:
- This federation effort is aimed at two-way data access between both databases, and will help large national initiatives like Canadian Consortium for Neurodegeneration and Aging (CCNA).
- NDAR - Brain-CODE Federation:
- OBI and the National Database of Autism Research (NDAR) are working towards a federation effort which aims to provide two-way data access to researchers through both the Brain-CODE portal and the NDAR database.