Genomorama: genome visualization and analysis

Abstract

Background: The ability to visualize genomic features and design experimental assays that can target specific regions of a genome is essential for modern biology. To assist in these tasks, we present Genomorama, a software program for interactively displaying multiple genomes and identifying potential DNA hybridization sites for assay design.

Results: Useful features of Genomorama include genome search by DNA hybridization (probe binding and PCR amplification), efficient multi-scale display and manipulation of multiple genomes, support for many genome file types and the ability to search for and retrieve data from the National Center for Biotechnology Information (NCBI) Entrez server.

Conclusion: Genomorama provides an efficient computational platform for visualizing and analyzing multiple genomes.

Figure 1

Genomorama can load and display the multiple annotated contigs stored in a whole genome shotgun GBK file. This screen shot shows five contigs from Sphingopyxis alaskensis RB2256 (extracted from the NCBI [21] file wgs.AAIP.1.gbff) and the associated sequence quality scores (from the NCBI [21] file wgs.AAIP.1.qscore). Quality scores are proportional to the negative log of the probability that a given base has been incorrectly assigned as an A, T, G or C and are shown as black plots superimposed over each contig track. The value of a quality score for each track is interactively displayed on the menu bar as a user specified score [i.e. "user(90)"] for the annotation track and base currently selected by the cursor.

Figure 2

Comparing the time to load human chromosome 1. The time to load Homo sapiens chromosome 1 is used to compare the performance of Genomorama and two Java based tools: Apollo [2] and Argo [3]. The time to load the GBK file [GenBank:NC_000001.9] from the local hard drive is shown for three computing platforms: a high-end OS X 10.4.8 workstation (dual 3 Ghz Intel Xeon CPUs, 3 GB ram, Java 1.5.0), a mid-range Linux Red Hat 4.0.1 workstation (dual 2.4 GHz Intel Xeon CPUs, 1 GB ram, Java 1.4.2) and low-end OS X 10.3.9 desktop (single 1.8 GHz G5 PowerPC CPU, 512 MB ram, Java 1.4.2). The Java-based programs were run from the command line with the arguments "-Xms32m -Xmx1024m" to increase the amount of memory allowed to the Java virtual machine. Providing Java with more than 1 GB of memory did not improve performance (results not shown). Each program loaded the genome file twice (to ensure fair OS disk caching) and the second load time is reported. For all platforms, Genomorama loads the genome file more than an order of magnitude faster than either of the Java-based programs.

Figure 3

Genomorama supports sequence searching with PCR primers. The genomic neighborhood of the amplicon (shown in orange) produced by the B. anthracis [GenBank:NC_003997.3] chromosomal specific PCR primers, M.Ctg032 [32]. The amplicon is contained within a glycosyl transferase (show in yellow). The amplicon annotation was added to the genome by selecting the "annotate" button on the Hybridize dialog box.