HRGRN: A Graph Search-Empowered Integrative Database of Arabidopsis Signaling Transduction, Metabolism and Gene Regulation Networks

The networks controlling plant signaling transduction, metabolism, and gene regulation are composed of tens of thousands of genes, compounds, proteins and RNAs as well as complicated interactions among them; these networks play critical roles in many plant fundamental functions, such as growth, development and environmental response. Of these networks, hormone related network has aroused extensive interests in research community due to its association with many vital biological mechanisms. 

However, the data regarding interactions among genes, proteins, compounds and RNAs scatters in published literatures, high-throughput datasets as well as third-party databases; and  many “unknown” yet important interactions need be mined and established through extensive computational analysis. In addition, these interactions are heterogeneous, i.e., various biological functions, and are difficult to be integrated. Thus, it is still challenging and time-consuming to explore the complicate gene interactions at genome-scale network level from existing heterogeneous interaction resources for biologists.

We developed HRGRN, a Graph search-empowered integrative database of Arabidopsis signaling transduction, metabolism, and gene regulation networks, with an emphasis on hormone signaling and gene regulatory pathways. The database is designed to host and discover interactions among genes, proteins, compounds, and small RNAs through integrative analysis of heterogeneous interactions.

The HRGRN utilizes the highly scalable, robust native graph database, Neo4j, to hosts and integrates biological interactions between proteins, compound and protein, transcription factor and target gene, small RNA and target gene, transporter and substrate, enzyme and substrate/product. We also added gene pair with similar expression pattern, gene family/complex and member as well as modified protein and original type into HRGRN as interactions. These interactions were either validated by experiments or predicted by computational approach. The hormone-associated interactions are highlighted to facilitate the exploration to hormone biosynthesis, secretion, signaling and regulation related functions in Arabidopsis.

Furthermore, the HRGRN integrates a series of built-in graph path search algorithms to enable and empower users to discover novel relationships between genes or pathways which might be missed by traditional search methods. Users can also build sub-network based on known interaction knowledge. It is also worth mentioning that the analysis results of HRGRN are visualized with text, tables, figures and interactive Cytoscape-based web interfaces. 

The HrGRN database is publically and freely available at http://plantgrn.noble.org/hrgrn/