What is RNAiDB?
This site provides comprehensive access to publicly available RNAi phenotypic data from C. elegans. RNAiDB provides raw data (in the form of images and movies), annotated phenotypes, graphical gene maps, analysis of potential off-target gene inhibition, and tools for searching and mining phenotypic data.
As of v3.0, RNAiDB contains all published RNAi experiments in C. elegans that have been deposited in WormBase, including data from the literature and published large-scale RNAi studies. RNAi to gene mappings for all experiments have been re-analyzed using ePCR and/or a sliding n-mer window method to identify all genes in different genomic locations that may potentially be inhibited by each experiment. Gene maps showing canonical and putative alternate mappings are displayed graphically on RNAi Experiment and Gene/ORF card pages.
Citing RNAiDB: To cite RNAiDB, please use this article from the 2004 NAR Database issue:
RNAiDB and PhenoBlast: Web tools for genome-wide phenotypic mapping projects.
Gunsalus KC, Yueh WC, MacMenamin P, Piano F. Nucl Acids Res (2004) 32,Database issue:D406-D410.
Full Text at NAR
Abstract at PubMed
How does RNAiDB work?
RNAiDB is powered primarily by an object-oriented database called AceDB, a database designed specifically for handling complex biological data. This means the data is divided, or "encapsulated" into named objects. These objects are created from templates called "classes". So, each piece of data in RNAiDB is of a specific type (class), and has a unique name to identify it. For example, there exists an RNAi Experiment class in our AceDB, and every RNAi Experiment ( e.g. PF:ORC_2G1 or Simmer:AH6.5 ) is created using this class. Most of our classes are broadly similar to those used at WormBase and NCBI's WormGenes , which are also powered by AceDB.
In addition to AceDB, RNAiDB also uses Sanger's General Feature Format ( GFF ) data via a relational database ( we use the open-source database engine MySQL ). This GFF data gives us coordinate-based positional information about features existing within the C. elegans genome and allows us to represent positional information in gene graphs.
RNAiDB was conceived by Fabio Piano while a postdoc in the laboratory of Ken Kemphues at Cornell University as a way to provide open access to RNAi phenotypic data in C. elegans. Fabio continues to provide cogent advice regarding further development of the database and has sponsored RNAiDB since moving to NYU.
RNAiDB v1.0 was constructed by Marco Mangone and Lincoln Stein from Cold Spring Harbor Laboratory in collaboration with Fabio Piano and was released in conjunction with Piano et al., 2000.
RNAiDB v2.0 (July 2002) was a major rewrite of v1.0 by Kris Gunsalus at New York University. Changes included revisions to data models, search capabilities, result summary page, gene card page; addition of RNAi Experiment card page; revised mapping of RNAi experiments to predicted genes using ePCR and BLAST; inclusion of new data on 751 germline-enriched genes described in Piano et al., 2002.
RNAiDB v3.0 (January 2004): RNAiDB was updated to include all available RNAi data from WormBase (version WS100), including data from all published large-scale RNAi studies in C. elegans. Potential off-target hits were annotated based on electronic PCR and BLAST similarity, and graphical gene maps are introduced to allow users to view the location of RNAi reagents and off-target mappings to the gene models. An array of new search features was introduced, including combinatorial searches and "PhenoBlast", a tool to allow the retrieval of genes with a similar phenotypic signature relative to a single query gene of interest. This database release was developed by Wan-Chen Yueh, Philip MacMenamin, and Kris Gunsalus and described in Gunsalus et al., 2004.
RNAiDB v4.0 (January 2005): RNAiDB was updated to include data from Fernandez et al., 2005 and all available RNAi data from WormBase (version WS135). The database was transferred to a new dual AMD Opteron server with 2.5TB RAID storage. A new sliding window method of annotating potential off-target hits was introduced based on contiguous stretches of matching 21-mers in pairwise alignments of RNAi reagents with spliced gene sequences (developed by Chuck Zegar). A relative score indicating the relative sequence similarity of the off-target hit to the canonical target was also created. Much of the underlying code base was revised to reflect new data models at WormBase and RNAiDB, to streamline the web site, and to simplify synchronization with future WormBase freezes. Primary developers: Philip MacMenamin and Kris Gunsalus.
RNAiDB v5.0 (January 2007): RNAiDB was updated to include data from Sönnichsen et al., 2005 and all available RNAi data from WormBase (version WS150). A new web interface and corresponding infrastructure was designed to improve maintainability and user experience, the sliding window annotations and PhenoBlast were re-implemented, and some new features were added (see RNAiDB News). The primary developer was Philip MacMenamin with database and curation support from Kris Gunsalus.