Analysis of Gene Relationships using a Graph Data Model

DESCRIPTION (Applicant's abstract): Genomic technology has become a major force facilitating biomedical research. Enormous data sets of great complexity are accumulating more rapidly than scientists can create specialized databases and analytical tools. The ability to synthesize and integrate disparate sources of genomics data into biologically meaningful information is a fundamental need. Commercially available products are not adequate in managing the data, mining the information for novel biological relationships, or elucidating components of biological pathways. Agilix Corporation will develop a novel algorithmic analysis method based on graph-theoretic tools, that will organize, catalogue, and data-mine genomic and proteomic relationships. The goal of this research is to develop a novel genomics analysis framework to unify heterogeneous genomics information into a common data structure and analyze gene relationships. In Phase I, we will: develop a comprehensive gene-graph model to organize and store gene relationship information as graphs; build methods to analyze and compare graphs; and create the software to visualize the gene-graph relationships. In Phase II, we will expand the number of graph structures and graph operators, and we will develop a local database and an analysis workflow engine. Phase III commercialization will include a low cost software application for the general research community. PROPOSED COMMERCIAL APPLICATION: Our new gene-graph technology will enable scientists to perform operations in several areas of bioinformatics previously inaccessible to those not trained extensively in bioinformatics. These areas include statistical analysis of gene expression data, data management, analysis workflow, data visualization, and the performance of "consensus" operations and simple mathematical operations such as addition, subtraction, and intersection on complex data sets. This new analysis capability will reduce the need for a team of bioinformaticians, and empower the biologist to synthesize and explore disparate genomic information at a level of sophistication not available in any other commercial package.