GeneTrack--a genomic data processing and visualization framework

Abstract

High-throughput 'ChIP-chip' and 'ChIP-seq' methodologies generate sufficiently large data sets that analysis poses significant informatics challenges, particularly for research groups with modest computational support. To address this challenge, we devised a software platform for storing, analyzing and visualizing high resolution genome-wide binding data. GeneTrack automates several steps of a typical data processing pipeline, including smoothing and peak detection, and facilitates dissemination of the results via the web. Our software is freely available via the Google Project Hosting environment at http://genetrack.googlecode.com

Fig. 1.

GeneTrack Composite Plot. Black vertical bars represent experimental sequencing reads that map to a genomic coordinate. The purple trace represents the smoothed fit over the read values. The purple horizontal tracks represent the 147 bp-wide nucleosome predictions as given by the peak detection algorithm. The blue and red tracks are yeast ORFs on the forward (blue) and reverse (red) strands.

Fig. 2.

GeneTrack Two Strand Plot. Within the software full strand information is maintained. The same plot from Figure 1 is now visualized on the separate strands forward (blue) and reverse (red). An ideal signal would be perfectly mirrored; this type of display allows one to identify systematic shifts in the data. Note how the fitting and peak prediction algorithms also operate on separate strands.