Comparative Genomics

[Introduction] [CompGen promoters and promoter sets] [Table Output] [Graphical Overview]

Introduction

The Comparative Genomics section in ElDorado allows analysis of the transcripts known for a group of orthologous genes (human, mouse, rat, chicken, dog, chimpanzee, rhesus macaque, cow, horse, pig, opossum, zebrafish and platypus). Incomplete or misleading annotation for one genome is identified by comparison of the information available from the other genomes. The program evaluates which of the available promoter regions correspond to each other and extends the annotation by additional promoter regions for which no transcripts are annotated so far (CompGen promoter). All of the promoter regions can be selected by the user for further analysis.

For the requested input gene all Locus IDs for a group of known orthologous genes and their alternative transcripts are retrieved from the database. The transcripts are sorted by organism and aligned by the most 5' conserved exon. Corresponding promoter regions from different organisms are grouped into promoter sets. The analysis is based on the comparison of the exon/intron structure of two transcripts and on the sequence similarity of the corresponding sequence regions.

For the assignment of orthologous genes all transcripts annotated in ElDorado are aligned against each other exhaustively. The pairwise sequence similarity is used to build homology groups covering up to 13 vertebrate genomes.

The output will show a table with all transcripts, a task menu for further analyses of the promoters and a graphical overview of the aligned transcripts.

CompGen promoters and promoter sets

We developed an algorithm to identify corresponding regulatory regions in the genomic sequence of different organisms which are responsible for the transcriptional control of orthologous genes.

Our approach correlates the transcripts known for a gene in two different organisms by mapping the first exon of a transcript in one organism to the genomic sequence of a second organism. If the mapped exon overlaps with the first exon of a transcript originally annotated in the second organism the two transcripts are defined as corresponding.

Background: Since the first exons are usually non-coding (5'UTR) they are not highly conserved between species. Thus, we defined potential target regions of several thousand base pairs for mapping in the second organism: The transcripts in both organisms were analyzed for conserved exons by comparing nucleotide sequence and length of the exons. Based on conserved 'anchors' the potential target region is determined (ca. 20.000 bp) and the first exons of the first organisms are mapped to the target region.

This approach was executed exhaustively for a large group of organisms for which orthologous genes have been identified before. The outcome of the genome wide analysis is a collection of promoter sets. Each of the promoter sets represents a group of regulatory sequences that are responsible for the transcription of corresponding transcripts from different organisms. Promoters in such a promoter set, that "only" have a transcript mapped from a different organism are called CompGen promoters.

Relevant transcripts

For comparative promoter analysis (e.g. FrameWorker analysis) it is important to use the correct promoters of a number of alternative transcripts as input. Genomatix marks transcripts that are probably not relevant for producing the main product of a gene as less relevant, thus allowing a focus on the main promoters for a gene for subsequent analysis. The tag less relevant is based on an evaluation of the exon/intron structure of all alternative transcripts within a locus.

Output

The result page contains three sections:

A table view of the transcripts

The first column ("Loci") lists the information about organism, chromosomal location, gene name, Genomatix locus ID, and GeneID for each of the loci. Following the links in this column either starts a complete ElDorado analysis or opens the More Gene Info for the respective locus. The contig accession number and strand specificity refer to the genomic sequence for this locus and the orientation of the locus as it is annotated in this genomic context. The - icon next to the organism name allows to hide the information for this locus. It can be re-displayed by clicking on the + icon.

The promoter regions belonging to the loci are shown in the next column ("Promoter Selection"). For each of the promoter regions the start and end position and the length of the promoter are provided. If the promoter region belongs to a promoter set conserved between the orthologous loci, the respective information is annotated. Promoters determined by the analysis of the available genome annotation from orthologous loci (CompGen promoter) are also listed. The check boxes in this column allow to select a set of promoter regions for sequence extraction or for further analysis.

The transcripts assigned to a promoter are listed in the last column ("Transcript"). For each transcript, the number of exons and the position of the TSS relative to the promoter region are given. The colored field in front of the transcripts indicates the quality of the transcript (gold, silver, bronze).
An analysis button to check for the presence of certain transcription factor binding sites in the promoters
When one or more transcription factor families (up to 5) are selected in the popup menu and the "Show" button is pressed, the page is reloaded and additional information is shown for each promoter: if ALL of the selected binding sites are present this is shown in red.

A task menu for further analyses

After selection of the desired length the selected promoters in the above result table can either be extracted or used to start an analysis with GEMS Launcher:

Save selected promoters allows to extract the selected promoters in either GenBank or FASTA format. The length of the promoter sequences can be defined by the user; up to 3,000 bps upstream and downstream of the annotated TSS are supported. The extracted promoter sequences are displayed in your browser window with the option to download them to your local hard disk or to copy them to your personal sequence directory.
If the Excel-format is selected, a tab-delimited file with all results from the table above and the promoter sequences is downloaded. It can be opened directly with Microsoft Excel.
"Search for common TF sites in multiple sequences" lets you step into the analysis of common transcription factor binding sites of the selected promoter sequences. The task will produce an output page with a graphical representation of the binding sites found.
To define a common framework of transcription factor binding sites based on the set of selected promoter sequences use the task "FrameWorker: Definition of common framework".
Potentially functional promoter modules can be found via the task "ModelInspector: Search for promoter modules" (using a highly specific library of predefined promoter modules).
To determine conserved sequence regions use "DiAlign: Local multiple alignment".
Transcription factor binding sites that are conserved within the alignment can be identified using "DiAlign TF: Multiple alignment plus TF sites".

Available tasks for selected promoters above

Select promoter length:
Genomatix optimized length
User defined length:

bp upstream of first TSS and

bp downstream of last TSS

(max. 3000)

Save selected promoters
in GenBank format
in FASTA format
export to EXCEL format

Submission of selected promoters to GEMS Analysis
Search for common TF sites in multiple sequences
FrameWorker: Definition of common framework
ModelInspector: Search for promoter modules
DiAlign: Local multiple alignment
DiAlign TF: Multiple alignment plus TF sites

A graphical overview of the aligned transcripts

The overview is a graphical representation of the transcripts listed in the table above and their genomic context. To allow an alignment of transcripts from different loci they are all displayed on the plus strand. Alternative transcripts from a single locus are simply arranged by their genomic location. Transcripts from orthologous loci are aligned against each other by their most 5' conserved exon. If there are no corresponding exons found the transcripts are aligned by their assumed TSS.

In this example (CAP12N gene) there are four transcripts available for the human genome. They are transcribed from three independent promoters. For the mouse genome there are only two annotations available corresponding to two transcripts from the human genome (promoter sets 2 and 3). For the rat genome only transcripts for the longer version known from the human genome are available (promoter set 2). Due to analysis of the information available for the three genomes the missing promoter region (promoter for promoter set 3 in rat) could be identified in the rat genome.

output

The Java graphic consists of four parts:

the main panel with the sequences and annotated elements (middle right)
a navigation panel with zoom- and scroll options (lower right)
a selection tree with elements to display or hide (left)
a toolbar with action buttons (top right)

For a detailed description of the general features of Genomatix graphics please check the graphics overview.

The main sequence panel

The main sequence panel contains all transcripts as listed in the table view of the output. For each transcript, a separate sequence is depicted as a line with all annotated elements. Thus, overlapping transcripts within one species appear below each other with differently annotated exons, introns or UTRs.

The colored field to the right of the transcript labels indicates the quality of the transcript (gold, silver, bronze).

The navigation panel

The view on the sequence can be changed by using the zoom- and scroll element in the lower right part of the graphics. The navigation panel contains a scaled down version of the sequence and a red box which marks the currently selected part of the sequence that is visible in the main sequence panel above. By default, the whole sequence is displayed.

slider To zoom in or out, click on the red box in the navigation panel and adjust the box by reducing the size via its handles (the small white boxes top left and bottom right of the red box). The sequence in the main panel will adjust to the selected window.

If you want to scroll along the sequence, move the red box within the navigation panel by sliding it with the mouse to the desired position. Alternatively you can click anywhere inside the scrollbar to jump to the desired destination.

The selection panel

The selection panel contains a tree with all elements that can be toggled on or off for display. Tree hierarchies can be shown or hidden by clicking the + or - symbols respectively. Checkboxes provide the possibility to toggle the visibility of

all or selected transcripts
all or selected promoter sets (orange boxes)
all types of sequences elements annotated within ElDorado

Transcripts are ordered by species so that all transcripts of a species can be switched off simultaneously with one click. Additionally, individual transcripts can be removed from the main sequence panel.

For a detailed description of the sequence elements within the selection panel see the description of the detailed graphics in ElDorado.

The toolbar

	exporting the complete result page to HTML (including all graphics in JPG format)
	exporting the graphics or a selected region to a certain format (JPG, PNG, TIFF, PNM), based on the current settings of zoom and element selection
	selection of a region which can then be exported
	recalculation of the layout, i.e. the graphics is displayed as it was set up by default