Berkeley Drosophila Transcription Network Project

Berkeley Quantitative Genome Browser User Manual

Back to TOC


The Preferences dialog window provides a way to control global settings. It is launched by following the File->Preferences menus or typing CTRL-N (or system equivalent). Clicking the "Okay" button applies the changes only within the current application process. Clicking the "Save" button will write the preferences to an xml file in the user's home directory called .bqgb_preferences.xml, which can be additionally modified via a text editor should the user prefer. The xml file is read every time BQGB is launched.

Many preferences can be set file-by-file and saved between launchings

The preferences window is currently divided into three tabbed panes: Files, Colors and Accessibility. Colors and Accessibility should be straightforward and won't be discussed further here. The settings under Files are a bit trickier.

Working from the top of Files pane down, the first set of controls is labelled "Data source file". The names here should refer to the data source files for which one intends to set a preference. It's certainly not necessary to set a preference! However, preferences can be useful for frequently used files. Whatever file is currently displayed here references all the settings below it in the Files pane.

Below the "Data source file" controls are the "Feature read filter" controls. Here one may stipulate a set of features to read, or exclude, from a file. The list of features is given as a series of semi-colon delimited terms. The terms need to appear exactly as the features appear in their source file. For example, in the figure above, only gene, mRNA and CDS are read into memory from the source GFF. While that source also contains, for example, exon features, exon features will not be read. What constitutes a feature in a file varies by file type. In an SGR file, there is but a single feature of score values, so filtering it is meaningless. In a GFF file, a feature is simply the values of column #3. In a BED file, the feature refers to the "name" attribute given in the "track" definition.

Location read filter is the next set of controls. It is currently disabled, but in a future version should limit reads to selected biological regions of the data.

Next, one can choose a default graph type. It is not necessary to define a default type. BQGB can make a good guess about the most appropriate graph type for the data, ie. when data is score, annotation, annotation and scores, sequence, etc.... The most likely non-default uses would be to (A) set a file to open in data density mode (a heat-map style histogram showing data rich regions) or (B) select a bar and line display. In the case of data density, one could have a file open in this mode, which is useful at low zoom levels, and the later switch to another graph type via the "Track Appearance" tool after zooming to a region of interest.

If using the bar and line display notation, more information will be required to define the relations within the file that are the "bars" and the "lines" of each glyph. This is done via the "Relations", the bottom most set of widgets in the file preferences pane. Currently, BQGB only parses GFF attributes for relations, ie. the terms entered here are the key terms to GFF attributes that link together separate records such as gene to mRNA and mRNA to CDS. The terms must exactly match the key term as it is in the GFF file. So, if one examined the GFF file used in the example above, they would find that gene records have an "ID" attribute containing a value that matches with the value of the "Parent" attribute in associated mRNA records. These relations may be set for 2-tier and 3-tier relations to allow for alternate transcripts

Back to TOC