The complete genome can be visualized by CGView, the Circular Genome Viewer (see website, see PubMed: 15479716). This is an excellent tool developed at the University of Alberta, Canada, for zooming in and out of the circular genome structure.

For the metabolic network reconstruction we applied two different methods to identify the enzymes of the Pseudomonas species.
The enzymatic activity of proteins by the first method was determined in three steps:
1. The external databases KEGG, PGD, ENZYME, and BioCyc provide EC numbers for the proteins.
2. Homology analyses lead to orthologous protein pairs (see 'Comparative genomics').
3. If the sequence identity (calculated by the global alignment tool stretcher) equals or exceeds 60%, all EC number connections to proteins were transferred to their orthologous protein partner.

The second method was metaSHARK (see website, see PubMed: 15745999). EC-numbers were directly assigned to the corresponding enzymes by metaSHARK. The cut-off was set at 10e-20.

The EC number source on the website is indicated accordingly. EC numbers newly identified by our method are declared as 'predicted'.

The Genome structure of one gene along with corresponding gene orthologs is visualised with the Genome Explorer. This is provided with a hyperlink to the BRENDA website, on which homologous gene regions of organisms are visualised with a sequence identity above 60%. The tool is kindly provided by the BRENDA group and is located at the University of Cologne. Click here for further information.

The visualization of metabolic pathways from KEGG in SYSTOMONAS is based on GraphViz (see website) using the dot layout. All known metabolic reactions are depicted for one pathway. Rectangles depict metabolic reactions, ellipses represent metabolites whose names are abbreviated with an asterisk * when the length exceeds ten letters. Both types of nodes are clickable. Different colours for rectangles depict distinct Pseudomonas species, which catalyse the corresponding reaction. These pathways can be obtained from metabolic pathway entries. An abbreviation code for the species is provided with the visualization output (AO1=P. aeruginosa PAO1, A14=P. aeruginosa PA14, P=P. putida KT2440, Pf-5=P. fluorescens F5, F01=P. fluorescens PfO-1, ST=P. syringae pv tomato, SP=P. syringae pv phaseolicola, SS=P. syringae pv syringae).

Functional characterisation of well-annotated proteins can be used for the characterisation of homologous / orthologous proteins. Finding further metabolic enzymes in the proteome by homology is of specific interest for filling in gaps in the metabolic network. One approach of determining homologous proteins was using BLAST and by using stretcher of the EMBOSS package subsequently.

The local alignment tool BLAST delivered comfortably quick results, that need to be evaluated by the global sequence alignment tool stretcher. The BLAST-database consisted of all pseudomonad proteins in SYSTOMONAS; the e-value cut-off for the pairwise, reciprocal BLAST search was 10e-50. Application of stretcher on the BLAST-pairs served as a further quality control of the homologous protein pairs. Transfer of EC numbers to homologous protein-pairs in SYSTOMONAS are based on homologs, that share at least 60% sequence identity. The e-value of BLAST and the identity calculated by stretcher can be retrieved by activating the two multiple alignment tools MUSCLE or Jalview.

Nearly each gene and its corresponding protein of P. aeruginosa has been assigned to a unique identifier called Locus ID. Therefore, we provided the option to search with Locus IDs in gene and protein tables.

Jalview is developed at the University of Dundee and visualizes multiple alignments and phylogenetic trees (see website see PubMed: 14960472. Its functionality in SYSTOMONAS is provided with homologous proteins (see section 'Comparative genomics'). Protein names within Jalview are composed as follows: species abbreviation_systomonas accno_e-value of pairwise BLAST-alignment_identity of pairwise global alignment calculated by stretcher. For example PAO1_PR0003261_2e-72_61 reads as protein of P. aeruginosa, strain PAO1 with SYSTOMONAS accession number PR0003261, that has a BLAST e-value of 2*10e-72 and 61% identity to the compared sequence. As data needs to be prepared as calculated multiple alignments, data is preprocessed by MUSCLE (see website, see PubMed: 15034147) prior to Jalview.

MUSCLE, MUltiple Sequence Comparison by Log-Expectation, creates multiple alignments and phylogenetic trees (see website, see PubMed: 15034147). Its functionality in SYSTOMONAS is provided with homologous proteins (see section 'Comparative genomics').

The results of our metabolomics experiments on P. aeruginosa can be retrieved via the query form 'Metabolome Data'. Experiment conditions and methods are indicated along with the raw data. Data can be plotted against different conditions.

Without these databases SYSTOMONAS would not be existing:

Data from these external databases is liable to copyright regulation of the corresponding database. Whenever functional characterization was transferred to homologous / orthologous proteins, the source is indicated as 'predicted'. For identification of the related proteins see section 'Enzyme designation'.

Following abbreviations are used on the SYSTOMONAS table websites for Pseudomonas species and strains:

For storing data in databases it is inevitable to assign unique identifiers. These help to specifically refer to one data entry. We provide unique accession numbers for pathways (PAxxxxxxx), interactions (INxxxxxxx), compounds (CPxxxxxxx), proteins (PRxxxxxxx), and genes (GExxxxxxx).

Virtual Footprint predicts transcription factor binding sites and deducts corresponding regulons. We adapted this tool to SYSTOMONAS by limiting the analysis on Pseudomonas organisms. Virtual Footprint is kindly provided by the PRODORIC group. Click here for further information.