20 Aug 2002

Gene duplications have been associated with a variety of genetic disorders (see Lupski for review), making their identification and characterization a medical priority. At first blush, the complete sequencing of the human genome should make this task easier, but in fact, the high degree of sequence similarity between duplicated segments and their original alleles makes it challenging. In the August 9 Science, Evan Eichler, Case Western Reserve University, Cleveland, Ohio, and colleagues at Celera Genomics in Rockville, Maryland, reported a method to detect such duplications that should facilitate both the identification of deleterious gene rearrangements and a more complete annotation of the genome.

First author Jeffrey Bailey and coworkers searched for duplications by looking at how many times a given gene was represented within the complete genome, rather than looking for sequence differences. They reasoned that overrepresented genes are more likely to have been duplicated. Using this approach the authors found in a whole genome shotgun sample 169 regions that were flanked by duplications, which the authors call rearrangement “hot spots.” Of these hot spots, comprising almost 300 megabases of DNA, 24 are associated with known genetic disorders including Gaucher’s disease, muscular dystrophy, and neurofibromatosis.

The authors search was limited to gene segments showing greater than 90 percent identity, or in other words genes that duplicated and diverged within the last 40 million years or so. In this regard the data may shine some light on what genetic changes have resulted in the separation of human and primate lineages. But perhaps more importantly, analysis of the remaining 145 hot spots may reveal some new genes that are associated with hereditary diseases.—Tom Fagan