Skip to main content
ARS Home » Research » Publications at this Location » Publication #175491


item Lazo, Gerard

Submitted to: Meeting Abstract
Publication Type: Proceedings
Publication Acceptance Date: 5/1/2004
Publication Date: 6/4/2004
Citation: Lazo, G.R. 2004. Genome visualization through gene expression. Alabama A&M Biotechnology Symposium Proceedings. p. 20.

Interpretive Summary: A presentation was provided to illustrate new data-mining tools used to monitor gene expression profiles. The tools primarily demonstrated utilized annotations attached to expressed gene sequences, but was also able to make use of high-throughput methodologies common in the laboratory. The survey of tools focused primarily on showcasing a visualization tool called Contig Constellation Viewer.

Technical Abstract: A major focus is placed on understanding gene expression considering the availability of completely sequenced genomes. High-throughput methodologies, such as DNA microarrays, provide opportunities to survey large volumes of data points at a time. Methods have been developed to assist the tracking down and identification of key differentially expressed genes. Because DNA microarrays rely primarily on genome sequences and expressed gene information, tools were developed to best make use of oligonucleotide information. Probe design attempts to represent as many genes as would be expected in a genome, and these numbers rely on evaluations using clustering algorithms applied to non-redundant sets of expressed genes. Important to this effort is the wealth of annotations and sequence information associated with EST collections. A visualization tool, named Contig Constellation Viewer, was created to view cluster relationships within the realm of attributes and properties of the libraries used as part of the clustering procedures. The individual ESTs, representing mRNA fragments, are assembled into contigs to predict gene structure. The visualization tool allowed a global view of all the assembled contigs and differentiated them from one another by determining the contribution of libraries to the contigs, and sorted them based on library properties. A survey of the assembly allowed distiguishing genes which were apparently library-specific from those which were constitutively represented.