The development of the genome annotation system GenDB started in 1999 when the first whole genome sequencing projects were initiated by the chair of genetics at Bielefeld University. Over the past years the system has been continuously adapted to changing requirements and numerous new features were implemented to support microbial genome projects in the best possible way. Recently, we have extended the data model of the system and developed the necessary analysis and visualization tools to handle eukaryotic genomes as well. Some of the main features of the GenDB genome annotation platform are briefly listed below:
- web-based interface to allow users from all over the world to access a GenDB installation
- flexible configuration system for customizing analysis
- support for various gene prediction tools (Critica, Glimmer, Prodigal, Gismo) and as well for other kind of regions (tRNAScan, RNAMMER, TransTerm etc.)
- BLAST, HMMER 2/3, InterProScan and various other tools for function prediction
- automated pipeline to process from raw contig to fully functional annotated genome
- emphasis on both manual and automatic annotation
- visualization and browsing of genome using various classification schemes, e.g. EC number and COG functional categories
- role-based access management
- complete automated processing of an E.coli sized genome within a day
- ability to interact with other software developed in our group, e.g. EDGAR
GenDB has already been used in the annotation of over 60 bacterial genomes. Support for eukaryotic genomes is available on request based on the GenDBe platform that was recently developed for analyzing the genomes of the Chinese Hamster, sugar beet and other crop plants.
The following pages illustrate the main features of the GenDB software platform.
For accessing a GenDB project enter your user name and password on the login page.
After the login, a user may select the GenDB project he or she wants to work with. A user may have access to any number of projects, and for each project the project leader assigns a role to the user, granting access to different functionalities offered by the GenDB web interface. In the example shown below the user has access to the "GenDB-2.1-Test" project and has been assigned the role "Chief":
After selecting one of the available projects, the annotation of a genome can be carried out in basically 6 steps:
- Selecting a contig
- Selecting a region to be annotated
- Analyzing the observations
- Editing the region
- Annotating the region
The GenDB main window is the central starting point for browsing a genome. After the project selection, the main window of GenDB pops up, showing a representation of the first contig of the selected project (Contig view). Arrows mark the positions of the gene predicted and annotated by the GenDB system. Their position also reflex their coding frame, with the three possible frame on the forward strand shown above the center line, and the three possible frames of the reverse strand shown below it. The white lines under the arrows represents the individual gene prediction result of the various tools involved in the gene prediction.
The arrows' colors indicate the various states of regions in the GenDB system and are freely configurable in the preferences. The user may switches between various color schemes for function annotation state, classification or regional annotation state. GenDB emphasis both on manual and automatically annotation, with the later being automatically started on data import. The user may decide which genes he want to analysed further during manual annotation.
GenDB allows working with several contigs in the same project, e.g. in case of unfinished genomes. The Contigs menu in the upper menu bar allows the user to choose the contig to work with.
All results computed by the tools integrated in GenDB (e.g., BLAST, HMMER, TMHMM, SignalP, etc) for the selected region are shown in the Observations window. In the main table of his window the tool results are sorted from "good" to "bad", according to the e-value. The user can configure the order of the results, restrict the maximal number of observations to be shown according to a given scoring threshold, or display only the best observations resulting from each tool.
The Annotation dialog displays several entry fields containing all collected information that has been initially computed by the automatic function annotation tool Metanor. Based on the analyzed observations and the start codon of the selected region, the user can make modifications in the data, or fill in empty fields (where no automatic value has been assigned). The changed annotation is submitted to the system by clicking the button Add new, on the top right corner of the dialog. Afterwards, the data is submitted to public sequence databases (on request) - except for the "Comment" field's contents, where the user can write informal comments, only visible to project members.