Berkeley Drosophila Transcription Network Project


Berkeley Quantitative Genome Browser User Manual

Back to TOC

Introduction to BQGB

The Berkeley Quantitative Genome Browser (BQGB and also known as "bbrowser" when executing the application) is a tool for visualizing and perusing linearly arranged data, ie. genomic and similar codes. The browser assumes that its input data uses nucleic base pairs as an abscissa; although strictly speaking, nothing enforces this requirement and BQGB could be used to view, for instance, amino acid sequences or time series data. The ordinate may be absent altogether, such as with gene annotations, or it may by numeric data associated with a single position or spanning a region. As BQGB's name suggests, the application tries to do a good job handling quantitative data and is moving toward an increasingly rich set of tools for handling numbers.

Unlike many genome browsers, BQGB is not web based. This means data like annotations, sequences and other numeric mark-up exist in files local to the machine running BQGB. Removing internet transactions is generally a great aid in increasing navigation speed as long as one is careful about not exhausting their machine's memory, which certainly is possible with many genomic annotations (see comment in paragraph below about pre-filtering files with grep, awk or similar tools). Many web based browsers sit on top of a specific database and contain detailed rules constraining how that data may be used. This is a safe but limiting practice. It is limiting when working with large amounts of custom data, since this data must be uploaded to a central server. And, it is limiting when working with numeric data which may see many numerous transformations in the course of browsing. Since BQGB is geared towards custom and numeric data, it is assumed the user is ready for a certain level of archiving and pre-processing responsibility for their own data! One should, of course, know what is in their data set and not assume that BQGB, unlike some more contraint filled browsers, will stop certain comparisons. For example, nothing stops one from simultaneously loading yeast, fly and human data. While normally the browser will only display a single chromosome name at a time (ex. Dmel chromosome 2L), the only check is on the name of the chromosome!2

Working with files on local disks and in memory can, in many cases, greatly boost performance. However, some datasets are just too large. While the browser provides filtering tools, again, it is hardly a replacement for time trusted, fast and simple tools like sed, grep. It is often faster, easier and more computationally efficient to use these tools to trim files to relevant data before loading them into the browser's memory.

BQGB is provided freely along with its source code. Please see the license distributed with the download for additional legal information; pretty typical open-source wording. The implementation language is C++ with extensive use of Trolltech's Qt libraries, which carry additional legal constraints for their reuse in commercial software. User comments, suggestions and bug reports are greatly appreciated, even when they can't all be immediately incorporated.

Thanks for checking us out!

Back to TOC