Pevzner and Tesler Co-author Landmark Papers in 'Nature' and 'Genome Research' Journals

December 4, 2002

A leading expert in computational biology at UCSD estimates that it took many more evolutionary genome rearrangements than previously thought--both large and small--to account for differences in the human and mouse genomes. The findings of computer science and engineering professor Pavel Pevzner and project scientist Glenn Tesler are included in two landmark papers recently announced. In the Dec. 5 issue of the journal Nature, Pevzner and other scientists in the 31-institution Mouse Genome Sequencing Consortium published a near-final genetic blueprint of a mouse, together with the first comparative analysis of the mouse and human genomes. In a companion paper published in the Genome Research journal, Pevzner and Tesler (in collaboration with Michael Kamal and Eric Lander at the Whitehead/MIT Center for Genome Research) analyze human-mouse genome rearrangements for insights about the evolution of mammals, and outline their development of a new algorithm to differentiate macro- and micro-level genome rearrangements.

Their conclusion: although the mouse and human genomes are very similar, genome rearrangements occurred more commonly than previously believed, accounting for the evolutionary distance between human and mouse from a common ancestor 75 million years ago. "The human and mouse genome sequences can be viewed as two decks of cards obtained by re-shuffling from a master deck--an ancestral mammalian genome," said Pevzner. "And in addition to the major rearrangements that shuffle large chunks of the gene pool, our research confirmed another process that shuffles only small chunks." "We now estimate over 245 major rearrangements that represent dramatic evolutionary events," added Tesler. "In addition, many of those segments reveal multiple micro-rearrangements, over 3,000 within these major blocks-a much higher figure than previously thought (even though some of them may be caused by inaccuracies in the draft sequences)."

Biologists and medical doctors study two kinds of rearrangements: clinical and evolutionary. Clinical rearrangements are manifested as a rather common chromosomal abnormality, associated with such diseases as Down syndrome, cancer, and infertility. Many healthy individuals also carry an asymptomatic chromosomal rearrangement. Evolutionary genome rearrangements, on the other hand, have occurred only about once or twice every million years in the course of mammalian evolution. "While clinical rearrangements affect a single individual, evolutionary rearrangements affect all individuals in a particular species and lead to speciation, for example, to separation of human and mouse lineages," explained Pevzner. "The human and mouse genome sequences can be viewed as two decks of cards obtained by re-shuffling from a master deck--an ancestral mammalian genome."

Because the mouse carries virtually the same set of genes as the human but can be used in laboratory research, information about the mouse genome will allow scientists to test experimentally and learn more about the function of human genes, leading to better understanding of human disease and improved treatments and cures. Said Pevzner: "If clinical and evolutionary rearrangements are related (still an open question), then studying evolution could yield insights about cancer and infertility, and vice versa."

In the Nature paper, scientists comparing human and mouse genomes found that more than 90 percent of the mouse genome could be lined up with a region on the human genome. That is because the gene order in the two genomes is often preserved over large stretches, called 'conserved synteny.' In fact, the mouse genome could be parsed into some 350 segments, or chapters, for which there is a corresponding chapter in the human genome. For example, chromosome 3 of the mouse genome has chapters from human chromosomes 1, 3, 4, 8 and 13, and chromosome 16 of the mouse has chapters from human chromosomes 3, 21, 22 and 16. Although virtually all the human and mouse sequence can be aligned at the level of large chapters, only 40 percent of the mouse and the human sequences can be lined up at the level of sentences and words. Even within this 40 percent, there has been considerable editing, as evolution relentlessly tinkers with the genome. The change is so great in most places that only with very sensitive tools can scientists discern the relationships.

The high-quality draft (near-final) sequence of the mouse genome was assembled by the Mouse Genome Sequencing Consortium, an international team of scientists at four sequencing centers in the U.S. and Europe. Those centers were joined in the analysis effort by scientists from 27 institutions in six countries. These included scientists on five University of California campuses--Berkeley, Santa Cruz, Santa Barbara, San Francisco, and San Diego. "It is interesting to note that all UC participants in the Nature paper are bioinformaticians rather than 'wet bench' biologists," noted Pevzner. "This reflects a paradigm shift in biology: it is quickly turning into a computational science. UC, and UCSD in particular, have recently invested heavily in bioinformatics development to further strengthen their already very strong wet biology presence."

Among other findings reported in the Nature paper: The mouse and the human genomes each seem to contain in the neighborhood of 30,000 protein coding genes. And about 5 percent of the genome contains groups of DNA letters that are conserved between human and mouse. Because these DNA sequences have been preserved by evolution over tens of millions of years, scientists infer that they are functionally important and under some evolutionary selection.

Links
Pavel Pevzner Research Profile
Pavel Pevzner Home Page

Story Adapted from Jacobs School Press Release
Full story can be found in the Jacobs School News Room
Media Contact: Doug Ramsey, (858) 822-5825