It was quite a summer for the comparative genomics crowd. A number of papers came out — including one in Genome Research from comparative guru David Haussler; one in Science from a consortium of scientists led by Harris Lewin at the University of Illinois at Urbana-Champaign and Bill Murphy from Texas A&M University; and one in Nature from senior author Bob Waterston comparing human and chimp — with a seemingly continual stream of findings derived from aligning sequences or chromosomes from different species’ genomes.
Just a few years ago, this field was still on the fringe of mainstream genomics; now, it seems almost impossible to avoid. What happened? Adam Siepel, who led the study in Haussler’s lab, refers to the “comparative genomics explosion,” positing that the success of early comparisons, such as human and mouse, has won over the skeptics and charged the field toward more and more comparative work. Lewin says a track record of significant results and the generation of high-quality genomic data has not only spurred researchers on in the field, but has served as “a real reason for why we need to have higher-pass coverage of more genomes.”
Lewin, director of UIUC’s Institute for Genomic Biology, helped lead a team of researchers comparing chromosomal organization across eight mammalian species (including human, pig, rat, cow, cat, mouse, horse, and dog) to study breakpoint reusage. Using high-resolution maps and new visualization tools, the scientists were able “to show that cancer breakpoints overlap with evolutionary breakpoints in the human genome,” Lewin says, as well as to “compute the putative ancestral genome” of those organisms. One surprise, he says, was the frequency: “20 percent of the breakpoints were reuse breakpoints.” The team also noted “for the first time that telomeres are highly conserved in their positions as telomeres in several genomes,” he adds.
The study, completed over the course of two and a half years, “shows what happens when you put together a multidisciplinary team of talented people and go after a problem in a big way,” Lewin says. The collaboration — also comprised of people from NCI, NHGRI, the Genome Institute of Singapore, the University of California, San Diego, and the University of Rennes — will continue as researchers try to get breakpoint information for other species, as well as clean up some of the lower-resolution maps with sequence information, Lewin adds.
In the Haussler paper, scientists studied several species of vertebrates, insects, worms, and yeast to get a better understanding of conserved DNA sequences across types of organisms. The evolutionary distances between all of those organisms was too great to compare genome to genome, says Siepel, a graduate student in Haussler’s University of California, Santa Cruz, lab; rather, the team constructed an alignment of each set of organisms (all the vertebrates, for example) and then “compared those conserved sequences across groups.” What Siepel refers to as “the headline finding” is that as organism complexity diminishes, so does the rate of conservation in non-protein-coding sequences. While that wasn’t a surprising result, he says, “it allowed us to say quantitatively” what people already suspected. Another finding that did surprise him was “how much conserved noncoding sequence there was in insects, which people tend to believe are quite a bit less complex than vertebrates,” he adds.
The study relied heavily on phastCons, a new computational tool Siepel developed that takes sequence alignments and uses phylogenetic hidden Markov models “to find regions that are unlikely to be as conserved as they are under neutral [evolution],” Siepel says.
With new tools like phastCons and the increasing amount of sequence data, the comparative genomics field is enjoying what Siepel calls “a feedback loop” that will continue to give the community momentum for more and more of this kind of research.
— Meredith Salisbury