Connectome Coordination Facility
Gregory K. Farber, Ph.D.
Office of Technology Development and Coordination
The goal of this initiative is to: (1) maintain a central data repository for Human Connectome data; (2) create a helpdesk service to answer questions from investigators who are trying to collect data that are compatible with the existing Human Connectome data; and, (3) to serve, in a limited capacity, as a quality control for new data that are being deposited.
The initial phase of the Human Connectome Project (HCP) will be complete in September 2015. That program is funded by the Neuroscience Blueprint and managed by NIMH on behalf of the Blueprint. The initial phase of the program focused on the development of two new MRI scanners with gradients that are particularly well suited for diffusion imaging. Those awards were made to David Van Essen at Washington University and Bruce Rosen at Massachusetts General Hospital. Both awardees successfully completed their technology development effort. The Washington University award also includes a second phase that is focused on collecting imaging, behavioral, and genomic data on a cohort of 1200 normal subjects aged 22-35. The purpose of this study is to understand the natural variation in connectivity that exists in the population. The imaging and behavioral data from these subjects are being made available to the research community shortly after collection (http://www.humanconnectome.org/ ).
The initial results from the HCP suggest that the data are of very high quality and that correlations between the connectomes and the measured behavioral data are being discovered. The family structure of the study population (i.e., twin and non-twin family members) has been especially useful in finding these correlations. Because of the very high quality of the data and the data collection technology, the HCP dataset and imaging protocol offer a very rare opportunity for multiple laboratories to collect data that can be effectively aggregated from different cohorts.
This initiative would accomplish the following goals:
- Maintain the existing data repository infrastructure for Human Connectome data and expand to include Connectome data from other research laboratories.
- Create a help desk to advise the research community about the best data collection strategies that will allow harmonization between their new data and the existing data. It is expected that many of the questions will involve the details of using pulse sequences on the local MRIs.
- Serve, in a limited capacity, as a harmonization center where data from multiple labs are made comparable. If the helpdesk feature works as expected, this harmonization will likely focus on behavioral/clinical instruments that have not been used in the HCP program up to now. The harmonization center will also perform a limited quality control inspection on the images coming from other laboratories.