Searching and Mining

InnateDB is freely available to the public as a tool for innate immunity research and can be mined as a knowledgebase where users can search for particular genes or proteins of interest and their associated interactions and pathways. Alternatively, InnateDB can be mined in a more high-throughput fashion, where users can upload a list of genes and carry out batch searches of InnateDB. Batch-searching allows users to return all interactions or pathways associated with a list genes and to integrate their own gene-expression data to investigate differential gene expression in an interaction network or pathway context. All interactions may be downloaded in several formats including text-based formats (tab, csv, Excel), the simple interaction format (sif), and most importantly, the PSI-MI extensible markup language (XML) interchange format. Interaction networks may also be visualized in our Cerebral program, a java plugin for the Cytoscape network visualization software, which uses subcellular localization information to orientate interaction networks in more biologically intuitive pathway-like layouts. Our latest version of Cerebral allows the overlay of up-loaded gene expression data from up to ten different experimental conditions. InnateDB also has integrated human, mouse and bovine orthology predictions generated using our Ortholuge software. Ortholuge uses a phylogenetic distance-based method to identify possible paralogs in high-throughput orthology predictions. Integrated human, mouse and bovine conserved gene order and synteny information has also been determined to provide further support for orthology predictions. Orthology information can be used with our associated tools for cross-species analysis of innate immunity gene expression data. The InnateDB data structure is based on the BioPAX data model, which will allow us to provide curated pathway information in a BioPAX compliant manner in the future. Currently, pathway cross-references can be downloaded in other text-based formats.