COG version and BAG version

1. Introduction to COG version

OperonViz (Graphical Operon Visualization Tool) is a group of CGI programs to provides graphical visualization of gene neighborhoods. At present, two versions of OperonViz are embedded in PLATCOM system. One uses NCBI's COG system to identify homologs (OperonViz-COG) and the other uses Dr. Sun Kim's Clustering algorithms, called BAG, for the same purpose (OperonViz-BAG). This manual is for OperonViz-COG.

• For more information about COG system, please visit NCBI's COG website.
• For more information about the operon-finding approach used in OperonViz program, please refer to my presentation.

2. Service Web Interface

Fig 1. OperonViz Web Interface (Using COG)

PLATCOM website currently provides OperonViz service. Fig. 1 shows a service request form. Three parameters (1) Search pattern, (2) Axially-aligning COG and (3) Intergenic distance between 2 genes need to be submitted by users.

Fig 2. Parameters and Pattern Search

The grammar of search pattern is blooean expression. and / or / not and ( ) can be used. (e.g. COG1982and (COG0470 or COG0721) not COG0645). A given axially-aligning COG is used as the axis operon alignment and marked as a red colored-block.

3. Operon Visualization by OperonViz-COG

Fig.3 shows the result form. This form is divided into 2 parts; (1) Operon imagemap and (2) New search request form.

Fig 3. Service Request Result (Using COG)

Each block represents a gene and different colors mean different COGs. Red color is assigned to the centering component group(COG) axis and light-gray color is used to show genes whose COG numbers are not reported yet.

Fig 4. Operon (Gene Neighborhoods) Structure (imagemap)

when a gene is clicked, its corresponding, NCBI Entrez page is popped up as follows.

Fig. 5 NCBI's Protein Page

To search another match in the same genome set that is alreay chosen in the previous step, a new search request form needs to be submitted. To go back to the top page, click "Click to choose a new set of genomes" link.

Fig 6. New Search Form

OperonViz using BAG clustering algorithm

1. Introduction for BAG version

OperonViz (Graphical Operon Visualization Tool) is a group of CGI programs to provides graphical visualization of gene neighborhoods. At present, two versions of OperonViz are embedded in PLATCOM system. One uses NCBI's COG system to identify homologs (OperonViz-COG) and the other uses Dr. Sun Kim's a new clustering algorithms, called BAG, for the same purpose (OperonViz-BAG). This manual is for OperonViz-BAG.

BAG is a graph theoretic sequence clustering algorithm that utilizes two graph properties, biconnected components and articulation points. With the current prototype implemented in C++ with the LEDA package, we were able to compare many different sets of genomes. To use BAG, you need toget a license for LEDA . The license for academia is very cheap.

• For more information about BAG, please visit Dr. Sun Kim's Page. Dr. Sun Kim's Research Page.
• For more information about how to utilize these tools, please refer to my presentation.

2. Service Web Interface

Fig.1 OperonViz (BAG) web interface

This is a simple service request form. OperonViz input file (that is, "Modified" BAG result file) is required to run OperonViz-BAG. If you already have it, just fill out this form, upload the input file and then submit the form. If you don't have it, you need to use another tool, called BAG2OPeronViz. To generate an input file, click "BAG2OperonViz Tool" link.

Fig.2 BAG2OperonViz conversion tool

BAG program generates totally 6 files and BAG2OperonViz tool converts one of them (*.result) into OperonViz-BAG input file. Fill out the form and choose genomes you want to investigate. (at least 2).

Fig 3. Result page of BAG2OperonViz tool

Now you have an input file. Open it by clicking a given link (Read Direction) and save it as a text format. Go back to OperonViz service Request Form (or click the link at the bottom). Then upload the saved input file and submit form. (Fig.1)

Fig 4. An Image map of Gene Neighborhood

This is an image map. Each block represents a gene and different colors mean different homolog groups. Red color is assigned to the aligning axis. Whenever you click a gene in this image map, OperonViz redraws image map and the chosen gene and its homologs are marked with red color and whole gene neighborhoods are re-aligned by this new aligning axis.

Fig. 8 NCBI Entrez Page

Fig. 9 ClustalW Result

OperonViz (BAG) also provides a pop-up window to show brief information and direct NCBI link (Fig. 8) of proteins in aligning axis (marked as red blocks).You also run Clustal W by choosing genes from the table and submtting this form. Fig. 9 shows the ClustalW result.

Kwangmin Choi @ School of Informatics, Indiana University, Bloomington, Indiana, USA

Last updated on 4/1/2004