A Top-Down Approach to Whole Genome Visualization

The investigation of large DNA contigs like complete chromosomes or genomes requires novel methods of data visualization. The complex information contained in a genome, particularly the relation of its individual genetic elements, needs to be accessible in a comprehensive, intelligent and intelligible manner. The yeast genome is expected to contain more than 6,000 Open Reading Frames (ORFs). As yet, the function of many of these ORFs has not been characterized satisfactorily. Also, many ORFs are found to have redundant copies elsewhere in the genome that originated from common ancestors. Other genetic elements (e.g. Tss, delta-elements, t-RNAs) are present in multiple copies. To visualize these relationships, a top-down "genome browser" is introduced that enables inspection of genomic data at different levels of abstraction (e.g. chromosomes, coding/non-coding regions, high/low levels of similarity). This novel tool is a key component for the integrated services approach to biological sequence data management (Heumann et al. 1995) and is accessible through the world wide web (WWW). This work demonstrates how the genome browser visualizes the results of an all-against-all comparison of the elements in the yeast genome as a graph. Interactive navigational queries across yeast chromosomes along the lines of sequence similarity open versatile options for the detailed investigation of genome properties. For sequence comparison the hashed position tree HPT (Mewes & Heumann 1995) is applied. Sequence similarity relationships are represented using the genome similarity graph (GSG) (Heumann & Mewes 1996c).