Flexible genome–phenome queries, at the speed of thought
Seven Bridges Sonar enables researchers to intuitively query genotype and phenotype relationships within data. It allows exploratory data analysis, drives rapid hypothesis generation and facilitates discovery.
Whereas other solutions focus on providing visualization and search for single datasets, Sonar goes further to support exploration, refinement and hypothesis testing across data from multiple studies. Sonar actively empowers users to bring in their own data for integrated analysis.
By removing technical and logistical barriers, Sonar makes harmonized biomedical data readily available to translational researchers, empowering them to ask questions and obtain results interactively. Sonar also allows researchers to learn from data as it is accumulated.
This flexibility ensures Sonar is the platform where questions are rapidly and painlessly refined to hypothesis; these hypotheses are tested accurately and intelligently to provide answers.
Step 1: Exploration
Sonar allows deep and flexible exploration of data.
Exploring the data is as simple as asking a question—Sonar allows users to ask many questions. Questions take the form of queries within the genome, phenome and omics spaces. These queries allow logical combinations and can span multiple variants.
Step 2: Refinement
Sonar visualizations provide rapid feedback on cohort characteristics to the user.
This rapid feedback allows users to start with broad queries, see query result distributions, and drill down into more complex queries iteratively. For example, the cohort size informs researchers about the size of targetable market and power to detect associations. Phenotype and genotype data distributions can guide further refinement of the researcher’s query.
Researchers can use this insight into the data to build and compare case and control queries, allowing comparison of phenomic and genomic properties across these cohorts.
Step 3: Hypothesis testing
Finally, integrated analytic capabilities enable hypothesis testing across these cohorts.
Current methods include Kaplan–Meier survival analysis of the cohorts generated by different queries. Planned additions include genome-wide association studies (GWAS), Phenome-wide association studies (PheWAS) and other analysis tools.
This toolbox of analysis methods allows researchers to directly evaluate their initial hypothesis without leaving the Sonar platform. Examples include facilitating GWAS calculations across cohorts and user-customized analysis. The Sonar roadmap includes deeper integration with other Seven Bridges technologies to use advanced methods (such as protein folding and deep learning).
Precision medicine research built on proven technology
Sonar is complementary technology to the Seven Bridges Platform. Both products are built on the Seven Bridges Core Infrastructure, which also powers numerous other systems including the Cancer Genomics Cloud for the US National Cancer Institute and Cavatica for the Children’s Hospital of Philadelphia. The combination of Sonar and Platform will allow users to interrogate data, find differentiating factors, and optimize secondary processing.
Phenotype representation and search in Sonar is built upon the Seven Bridges Modeling and Data Integration Stack (SB-MDIS), which represents data in a schema-less triple store (RDF) graph database. This approach was chosen for its flexibility, and the increasing adoption of Semantic Web data modeling in the biomedical industry. This modeling can capture interrelations between studies adding to joint usefulness and representational power.