PathwayVoyager: pathway mapping using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database

Abstract

Background: Equally important and challenging as genome annotation, is the subsequent classification of predicted genes into their respective pathways. The Kyoto Encyclopedia of Genes and Genomes (KEGG) represents a database consisting of known genes and their respective biochemical functionalities. Although accessible online, analyses of multiple genes are time consuming and are not suitable for analyzing data sets that are proprietary.

Results: Presented here is a new software solution that utilizes the KEGG online database for pathway mapping of partial and whole prokaryotic genomes. PathwayVoyager retrieves user-defined subsets of the KEGG database and stores the data as local, blast-formatted databases. Previously selected datasets can be re-used, reducing run-time significantly. Whole or partial genomes can be automatically analyzed using NCBI's BlastP algorithm and ORFs with similarities below the user-defined threshold will be marked on pathway maps. Multiple gene hits are sorted by similarity. Since no sequence information is transmitted over the Internet, PathwayVoyager is an ideal solution for pathway mapping and reconstruction of confidential DNA sequence data.

Conclusion: PathwayVoyager represents an alternative approach to many already existing, more complex pathway reconstructions software solutions. This software does not require any dedicated hardware or software and is flexible and straightforward to use. It is ideally suited for environments where analyses on variable datasets are desired.

Figure 1

Pathway Voyager mapping procedure. The analysis and mapping procedure of PathwayVoyager is shown in a flowchart diagram. Manual selection of organisms and pathways present in the KEGG database, at the time of analysis, results in the retrieval of a specific set of protein sequences that are subsequently reformatted into a BlastP database. Protein query sequences are then used as templates for local BlastP analyses. Results are subjected to a user-defined threshold and subsequently parsed to retrieve tagged KEGG pathway maps. Pathway graphs and parsed BlastP results are stored as a flatfile database and can be displayed using the graphical browser. Opposite double arrows represent Internet-access using the KEGG API. Symbols used are according to general flowchart conventions.

Figure 2

PathwayVoyager main window. Screenshot of the graphical user interface. (A) and (B) indicate listboxes showing the organisms that can be individually selected and pathways represented in KEGG at the time of analysis. Button for exiting the software, redo the analysis, and accessing the setup are located beneath both boxes. Section (C) resembles the step-by-step design to start the analysis, retrieve the selected organisms and pathways, and subsequently obtain the respective protein sequences. In manual mode, the KEGG analysis and pathway retrieval can be activated by the user. Existing KEGG analyses can be reviewed by accessing the lowest button. On-the-fly options, for re-using previous organism-pathway selections and respective BlastP results, are located in the top region of section (C).

Figure 3

Interactive KEGG Pathway display. The screenshot illustrates KEGG pathway mapping for the glycolysis/gluconeogenesis pathway using the predicted ORFeome of the GAMOLA annotated L. acidophilus NCFM genome as query template. The e-value threshold was set to 1e-10. Section (A) shows all pathways used for this analysis. Section (B) lists the marked entries. Section (C) represents the graphical representation of a selected pathway. Elements exhibiting BlastP hits below the selected threshold e-value are marked as red boxes. Section (D) shows the corresponding BlastP results, comprising the respective query protein designation, the corresponding KEGG hit, its amino-acid length, the BlastP-score, and e-value. Entries are sorted by ascending e-values.