Genomics pioneer joins EECS

By Sarah Yang, Public Affairs

Bioinformatics specialist Gene Myers is joining the Berkeley faculty William K. Geiger photo

30 October 2002 | More than two years after the landmark sequencing of the human genome, the computer whiz behind the algorithms used to decipher millions of pieces of the genetic material is coming to Berkeley.

Gene Myers, former vice president of Informatics Research at Celera Genomics in Rockville, Md., has been appointed a professor in the Department of Electrical Engineering & Computer Sciences, effective Jan. 1. As part of his work in computational biology, he will work closely with researchers at the campus’s Department of Molecular & Cell Biology, Lawrence Berkeley National Laboratory, and the Department of Energy’s Joint Genome Institute.

It’s a return to academia for Myers, who taught at the University of Arizona in Tucson for 17 years before the race to sequence the human genome inspired him to pack up for Celera in 1998. He will play an active role in Berkeley’s Center for Integrative Genomics, which brings together researchers interested in molecular evolution, biodiversity, and genomics, with a heavy emphasis on computation.

“Right now, seven different animal genomes have been sequenced and assembled, but that number is likely to shoot up to 50 to 100 in the next five years,” says Michael Levine, professor of molecular and cell biology and director of the center. “We’ll be mining a treasure trove of data, and we need to have a way to compare these genomes in order to understand the mechanisms of evolutionary change and innovation, to essentially understand the diversity of life on earth.”

Clever and insightful
In 1981, after earning his Ph.D. in computer science at the University of Colorado, Myers joined the faculty of the University of Arizona. While there he worked with Jim Weber, a geneticist at the Marshfield Clinic in Wisconsin, on a radical concept: sequencing the human genome by mathematically chopping the entire genome into random bits, sequencing those pieces, then reassembling them in correct order.

The technique, called “whole genome shotgun sequencing,” had been used successfully to sequence and assemble the genome of microbes, most notably the Haemophilus influenza bacteria in 1995. But skeptics rejected the idea that the technique could be used for animal genomes, which have far more repeating structures than microbial genomes. With so many bits and pieces that look alike, conventional wisdom held, a computer couldn’t possibly reassemble them in exactly the right order.

Scientists in the publicly funded Human Genome Project were instead using a divide-and-conquer approach by taking medium-size chunks of the genome and sequencing those sections one at a time. “It was a tedious method that involved many more steps and was much more expensive,” says Richard Karp, University Professor at Berkeley and a leader in modern theoretical computer science. “Myers was clever and insightful enough to develop a computer algorithm that resolved the concerns people had about using shotgun sequencing for mammalian genomes. He was so confident this would work that he took the plunge and joined Celera. It was a gutsy move.”

It was at Celera that Myers had free rein to test-run his algorithms.

The first key, Myers explains, was to modify the traditional shotgun protocol to deliver the genome bits in pairs separated by known distances of 2,000, 10,000, and 50,000 letters. The second key was to design algorithms that identified the non-repeating parts of the genome that are easy to reconstruct and then link those parts to pairs that bookend a repeat sequence. This self-checking system was built into the algorithm so the probability of reassembling any piece incorrectly was virtually zero.

“They all said we were nuts, but nobody could give me a logical reason why it wouldn’t work,” says Myers. “That’s when I knew we had a winner.”

Less than a year after joining Celera, Myers and a team of 10 researchers — including Gerald Rubin, professor of molecular and cell biology at Berkeley — proved that this refined version of shotgun sequencing worked. In 11 months they wrote 500,000 lines of code to sequence the genome of the Drosophila fruit fly. The research was dubbed “Best Paper of the Year” in 2000 by “Science” magazine.

With the sequencing of the human genome complete, Myers is now preparing himself for new frontiers in bioinformatics. “We’ve decoded the human genome,” he says. “Now I want to decode the cell.”