Study Cores
The HeartShare Data Translation Center consists of 4 cores: an Administrative and Outreach Core, a Cohort Core, a Data Portal Core, and a Data Management Core:
The Administrative and Outreach Core oversees HeartShare program operations, the HeartShare research skills development (training) program, and the deep phenotyping prospective study. A key directive of the Administrative/Outreach Core is to ensure rapid and broad data sharing of clinical, imaging, and molecular data to promote the use and dissemination of all data that results from the HeartShare program (from both the parent study and all ancillary studies). The Administrative/Outreach Core also interfaces with each of the cores, the NHLBI, and the HeartShare Steering Committee (SC), External Advisory Committee (EAC), and Observational Safety Monitoring Board (OSMB).
The Cohort Core is collecting and harmonizing data, images, and omics from previously conducted (extant) cardiovascular disease epidemiological cohorts, heart failure clinical trials, and heart failure registries. All data and images are being deposited into the NHLBI BioData Catalyst cloud resource for maximum data protection and dissemination to approved investigators for analysis. These data (including omics) and images are being analyzed using 2 approaches: (1) a hypothesis-driven approach to investigate pre-defined HFpEF subtypes and molecular pathways and (2) an unbiased approach (e.g., unsupervised machine learning) to identify novel HFpEF subtypes. Identified subtypes will then be validated and further studied in the prospective component of HeartShare.
Cohort Core Scientific Approach
The Data Portal Core serves as the hub for all data and images for HeartShare (from both existing datasets and the prospective, deep phenotyping component of the study) in concert with BioData Catalyst. The Data Portal Core is also responsible for supervising collection of EHR data from the 6 clinical centers and developing and maintaining the patient-facing Eureka HeartShare app, which is described in detail below.
The Data Management Core oversees and ensures data quality, integration and harmonization of various data types, performs advanced analytics, coordinates biospecimen and imaging repositories and supports integration with the NHLBI TOPMed program for omics analyses.
Organization of HeartShare Data Translation Center
BioData Catalyst
BioData Catalyst is the NHLBI’s cloud-based platform for secure data sharing and analysis. It consists of a data ecosystem that is designed to be flexible, modular, and user-friendly, and contains 3 main components: Terra (genomic data), Seven Bridges (analysis workspaces), and Gen 3 (publicly available data). All HeartShare data and images are deposited into cloud buckets (e.g., Amazon Web Services [AWS]) that are connected to Seven Bridges analysis workspaces. The HeartShare Data Translation Center serves as the data steward and controls access and permissions for all users of HeartShare data in BioData Catalyst. The figure below displays the flow of data and images from the various components of HeartShare into Seven Bridges where approved investigators can analyze HeartShare data and images.
Data Flow from HeartShare Components to Seven Bridges in BioData Catalyst
Access to Data
We are in process of assembling data and data sharing protocols. These protocols will be posted here.
For interested investigators, please check back soon, or email us at heartsharestudy@northwestern.edu for more information.