Berkeley Drosophila Transcription Network Project


Berkeley Quantitative Genome Browser User Manual

Back to TOC

Tools

Overview

BQGB's first generation of tools focus on navigation and track appearence, not data manipulation. Tools are located in tabbed dialog boxes/windows and in tool bars.

Tool boxes
Tool boxes

While the defaults open all toolboxes in a fixed position to the right side of the main window, tool boxes may be closed, dragged via mouse to the left side of the screen or left floating above the window. Tool bars default to closed upon application start-up.

Tool bar

Both tool bars and tool boxes may be toggled in and out of view by right clicking (or your system's equivalent) anywhere on the uppermost menu bar and selecting the desired tool, as shown in the screenshot below.

A right mouse click or equivalent on the menu bar will produce a drop down menu controlling the visibility of tool boxes and tool bars


Zooming and scrolling, marking and jumping<

Zooming and scrolling are mostly performed via the zoom and scroll sliders near the bottom of the central display area. In addition to mouse dragging these widgets, when the graph has keyboard focus, it may be zoomed and scrolled via the keyboard arrow keys.

The zoom cursor, a scroll bar near the top of the central display area highlighted in the image below, defines the point around which zooming will occur. This cursor also gives the point recorded by the "Mark" event and may be jumped back to using the "GoTo" button in the "Mark/Jump" tool bar. Double clicking on the zoom cursor toggles the cursor to show as a dotted line crossing all open tracks.

Zooming and scrolling can be controlled by the slider widgets circled above or by the keyboard arrow keys when the graph (or sliders) have the keyboard focus.


The sequence selector

When examining sequences that share a common biological alignment space, BQGB generally limits the view to a single sequence identification at a time, ie. a single sequence identifying a chromosome, plasmid, etc.... The sequence selector tool allows toggling between the available sequences. If a sequence is only available in some of the loaded files, only those files will show data tracks after toggling. This is a fairly firm rule with two exceptions: (1) a synonym has been set for a pair of sequence names and (2) when viewing multi-alignments where regions may map across different named sequences.

The sequence selector and navigation tree allows one to toggle between available sequences, set sequence name synonyms and set track visibility. Files may be closed via the tree navigator.

The sequence synonym setter offers a potentially useful, and dangerous, functionality. Its purpose was to simplify the comparison of sequences that have been named differently between files. In the screenshot above, one file named sequence 2L "2L" and another file used "chr2L". These two names have been set as synonyms. The danger with this tool lies in setting inappropriate synonyms, such as "2L" and "chr2R". A better strategy is to stick to a naming convention across files! Another possibility is to rename sequences inside a file before loading it into BQGB's memory using a simple tool like sed.

sed -i 's/chr2L/2L/g' some_data.sgr



The navigation tree

In addition to the sequence selector, the "Sequence Selector" tool provides a schematic of available data in the navigation tree. This tree parallels the organization of the underlying data structure with second level nodes always showing the file with which data is associated. The leaves of this tree correspond to tracks in the viewing area. The visibility of tracks can also be toggled from the navigation tree. The pop-up menu shown in the screenshot above is activated by a right mouse-click (or system equivalent). This menu also provides a switch for closing file and removing from their data from BQGB's memory. At this time, it is not possible to partially close, or clear from memory, parts of a file such as a feature or certain sequences.



Search/Filter

Two score tracks have been selected (double mouse-click on a track) and then filtered for values greater than 1.5. With the highlight function toggled on, all values are drawn with values above the threshold shown in the highlight color (red).

Searching and filtering are closely related operations. The criteria definitions are identical. In a search operation, the viewing area will be positioned at the next matching location in a the selected tracks (tracks with a dark outline selected via a double mouse-click). Filtering will not move the view area coordinates but will affect the visibility of the data in selected tracks, either masking-out data that fails search criteria thresholds or highlighting that data if the highlight checkbox has been toggled on.

The Search/Filter tool will only search/filter in selected tracks, although in the case of searcing, all tracks will move together to the position of a position match as expected. In the case of GFF files, a text based search may be further refined to only certain attributes such as "Name" or "ID".

Perhaps most powerfully, and also a potential for trouble, a search/filter operation may be composed of multiple named criteria across multiple tracks. Criteria may be logically combined with each other and tracks may be logically combined. For example, two criteria can be created with the "Add criteria" button and subsequently named by double mouse-clicking on the search criteria table's name cell. These two criteria might be combined with a logical or such as "select data greater than 1.5 (criteria #1) and less than 3.0 (criteria #2)". Tracks can be similarly combined. Let's say there is a single criteria, "select data greater than 1.5", and two selected score tracks. In this case a logical and between tracks means, "select only points that are shared between the two tracks and are greater than 1.5". A logical or means "select data that is greater than 1.5 in either graph and, if there are data in other selected graphs with the same coordinate, show it whether or not it meets criteria threshold". As one might imagine, using both criteria combinations and track combinations empowers the user to make their own logic errors, ie. the user has created a little program in the Search/Filter tool that may not operate as the user's internal logic expects. Of course, if one is sure their logic is correct and the results look amiss, please report the erroe as a bug! It always pays to independently verify unexpected results.



Track appearance

The track appearance tools allow textual manipulation of view space coordinates and graphical manipulation of tracks independent of defaults set in the preferences.

The track and appearance tool is a collection of functions affecting display position, visibility and color, separated into three tabbed panes. In general, track appearance controls only affect selected graphs (double mouse-click on graph area), which are marked by a dark border as shown in the adjacent screenshot. X-Axis coordinates are the sole exception since all tracks, selected or not, share a common abscissa.

The "Graph" tab provides a mechanism for changing graph types. Obviously, not all graph types are appropriate for all data, ex. annotations under "Score". Also, one will probably have better luck not toggling between "Bar and Line" representations and other graph types since this type places the equivalent of multiple tracks of data into a single visual track. Close the file and open its contents as a different track type (see Setting Preferences). Hopefully this behavior will be improved in a future release!

The "Visibility/Alignment" tag allows visibility toggling of a Y-Axis. Double clicking on the Y-Axis itself, followed by a mouse drag, allows the Y-Axis glyph to be moved to new positions in the track.

Finally, the "Color" tab sets the color for the graph. This provides a track-level level of control beyond setting color in the preferences, which would affect all tracks globally. The color setting, however, is not currently saved when the application is closed.



Info

An Info window showing GFF attributes and sequence data for the D. melanogaster gene CG2813.

The contents of the "Info" window vary depending on the graph type and source format of the data (ex. GFF versus BED). Info is drilled down to by clicking on features withing the track, this behavior varying by graph type. In the figure above, the source file is a GFF file and the "info" shown is largely taken from the GFF's attributes field. In this example, there is also sequence data available from an associated fasta file. The sequence data shows as a separate track above the annotation track that gives GC content by octomers, since zoom level is too far out to draw characters on the screen. The sequence itself, however, is given in the "Info" tab pane. If more than one sequence file were open, multiple sequences would be given, BQGB simply applying the feature's coordinates to a query of the available sequences.



Back to TOC