Probe Set Annotation

Annotation associated with the currently loaded probe set is displayed by the 'Probe Data' window which can be toggled using the button below the 'Mean Plot' button in the main window. This window displays information in 6 different widgets (sub windows) which can be toggled using the buttons at the top of the window. These widgets show the following information:
  1. Affymetrix: This displays annotation obtained from Affymetrix with the purchase of the chips. The description and probe set id are displayed in addition to a descriptive field I obtained from TIGR (The Institute for Genome Research) a long time ago.
  2. UniGene: This displays the UniGene gene, title and index fields for unigene entries linked to by the primary Annotation. Occasionally there is infomation present from more than one UniGene entry, this is a result of the original UniGene Affymetrix linked to having been split into more than one cluster.
  3. Genomic Alignments: Displays blast matches of probe set sequences to genomic sequences with alignments of predicted genes.
  4. Region: Displays the results of genomic region queries. These are displayed in the same area and in the same manner as the Genomic Alignments above. (In essence this button only toggles between a region centered view and probe set centered views).
  5. User Annotation: Displays any user annotation entered for this probe set, as well as any user sessions to which the probe set belongs
  6. Experiments: Displays short descriptions of the different samples present in the database.
Probe Annotation
Figure 1. Default View of Probe Annotation: The different views are toggled using the top row of buttons. Right hand column shows short description of different samples, top row of widgets display simple textual information wheres the genomic alignments, along with Ensembl Annotation are shown underneath. The bottom row of selectors control the size of the genomic region which is downloaded, the initial part of the region which is viewed and a set of parameters determining whether or not a given blast match is displayed in the system. Changes in the 'RegionSize' and 'Window Size' only take effect upon loading a new probe set.

Most of the information displayed requires little explanation, however, the genomic alignments view has a number of hidden capacities which are explained below.

Genomic Alignments
Figure 2. The alignments annotated.

This window displays the set of genomic loci which are associated with the currently loaded probe set. In addition to displaying the gene structures located in this region, the window also displays all blast probe set matches associated with the locus. The window is interactive and allows the user to zoom in on specific features as well as obtain peptide, mRNA and genomic sequences from the underlying database. The interactivity makes heavy use of the three mouse buttons normally used in X11 (the middle button is replaced with Ctrl Shift for those without a middle button).

Chromosomal Region
 The Chromosomal region currently loaded is represented by the lower horizontal white line in the above picture, on top of which is located the "amazing sliding range selector". The sliding range selector indicates the limits of the currently viewed chromosomal region. The sliding range selector can be manipulated in three different ways.
  1. Left Mouse Button. Press the Left mouse button with the cursor positoned above the selector, and then move left or right. This changes the lower limit of the viewed reqion.
  2. Right Mouse button. This works the same as the left mouse button but changes the upper limit.
  3. Middle Mouse Button (or Ctrl+Left Mouse Button). This changes both the upper and lower limits and thus appears to slide the range selector along the range.
The size of the chromosomal region loaded from from the central database (loaded concurrently with data from additional probe sets) is controllod by the little spinbox in the bottom left hand corner of the widget labelled "Region Size", and the initially viewed fraction of this is set in the spinbox to the right of this labelled "Window Size". Currently the maximum size of the chromosomal region which can be loaded is set to 1 million base pairs (but this may change). Changing the Region Size or the Window Size parameters in the spinboxes only has an effect when loading a new probe set. The currently viewed region can also be zoomed in by left clicking close to (but not directly on) the line indicating the DNA sequence and selecting a region using a selection box?
Gene Features
 The different transcripts for the genes predicted by Ensembl are shown either above (for genes lying on the forward strand) or below (for genes transcribed in the reverse direction). All of the alternative transcripts predicted are shown along with their intron exon structures. The limits of the predicted exons are shown with a white rectangle, within which translated exons are shown in yellow, and untranslated exons (or parts of exons) are shown in purple (note that the Ensembl predictions for untranslated exons are frequently rather poor and there are probably lots of missed exons). One of the genes has its transcript limits displayed with a red box. This is the gene which the data base system currently predicts the current probe set is representing. This linkage is based on the positions and qualities of blast matches to the probe set sequence in the mouse genome. This prediction is not particularly good or clever, and may be wrong. The user is encouraged to look at the locus with some care to determine if the prediction is correct or incorrect. Your poor stressed out administrator welcomes reports of stupid guesses by the system and will try to improve the system as the trends become more obvious. When the View is initially loaded the Ensembl Annotation for the predicted gene is shown in the table below. The annotation for the different genes shown in the view can be viewed in the table simply by clicking with the left mouse button on the boxes representing the transcripts.
The mRNA and corresponding peptide sequences can be retrieved from the central database system. In order to retrieve sequences from specific transcripts, these have to be selected first. Clicking on a transcript with the right mouse button selects or deselects (if already selected) the mRNA sequence from that transcript and changes the colour scheme of the transcript. Clicking with the middle mouse button (or Ctrl+Left Mouse Button) selects or deselects the protein sequence for that transcript and changes the colour scheme as indicated in the above figure. Both the peptide and the transcript sequence can be selected simultaneously, and their selection is independent of each other.
Probe Set Matches
 The Blast probe set matches are indicated along the white line representing the DNA sequence. Forward strand features are shown above the strand and blast matches to the reverse strand are shown below. Matches to the currently selected probe set (the one for which expression data is shown) is indicated by having twice the height of the other match features. The colour of a probe set match indicates the random expectation value given by the blast algorithm, with red for very low (i.e. good) random expectations, and green for not such good ones  (though this will probably change in the future).
The currently viewed probe sets can be loaded into the main index (i.e. allowing their expression data to be viewed) by first selecting the chromosomal regions and strands of interest by clicking either just above or just below the white line representing the DNA strand, and then by right-clicking elsewhere in the widget and selecting save sequences from the popup-menu (more details further down).