ODG: A Graph Database Generator for Omics Data

The Omics Database Generator(ODG) is a database generator for –omics data, especially suited towards plant genomes. It processes and integrates genetic, genomic, transcriptomic, pathway, ontological, miRNA and interactomic data from multiple genome releases, accessions, and species. Data is further processed by BLAST, InterProScan and target prediction, and integrated with annotation from Gene Ontology and ENZYME. Imported data is normalized and integrated into a graph database, with standardized relationships and properties, simplifying queries across different datasets. Database administration knowledge is not required. This procedure makes advanced data mining and comparative genomics simpler. Conserved neighboring gene pairs and clusters across multiple chromosomes and species can be rapidly identified. Gene identifiers which have been renamed in new genome releases can be disambiguated by examining neighborhoods. The database is accessible by an included web interface as well as programmatically via REST.

As an example application of this database system, we identify closely located gene pairs expressed primarily in roots found in both of the nodulating species Glycine max and Medicago truncatula, where the genes are closely located in both species. From this list, conserved pairs which are also found in Arabidopsis thaliana roots are removed and the remaining pairs are examined for presence/absence against Phaseolus vulgaris and Lotus japonicus, for which expression data was not readily available and not included in the database-build for this example. The same query is repeated with neighboring triplets in order to identify highly conserved nodule-organism specific genes.