GenDBWiki/FeatureTable

GenDB Features

The open source genome annotation system GenDB is based on a relational database management system. Contig sequences can be imported and after predicting the coding sequences (CDS), a number of standard bioinformatics tools like Blast, Pfam, InterPro etc. can be run on these regions as a basis for the functional (manual) annotation. By storing only a minimal required subset of each tool result (e.g. only a short description and a computed score) and recomputing the complete result (e.g. an alignment) on demand, the storage requirements can be reduced enormously.

The software has an object-oriented application programmers interface (API) implemented in Perl which has been partially generated automatically with O2DBI. The latter creates an object-relational mapping onto SQL tables. GenDB has a web frontend and a Gtk user interface that creates dynamic visualizations and can be used for annotating genes based on the computed tool results.

The system features the concept of wizards which are software agents that automate complex repetitive tasks (e.g.~ORF editor, frame-shift correction, contig update, etc.). In addition to a search interface, the software can generate statistical plots and includes a virtual 2D gel. The data contained in \gendb{} can be exported into widely used output formats like \fasta, EMBL, GFF (genome feature format\footnote{\URL[1]{/Software/formats/GFF}}), and others. An integrated pathway module supports the analysis and visualization of annotated enzymes based on the KEGG~\cite{kegg:2000} metabolic pathways and the PathFinder~\cite{PATHFINDER} software.