New Process Helps Scientists Sort Crucial Genes

Scientists from UCLA and a small Westwood biotechnology company are reporting a new, computerized method to sort through thousands of newly discovered genes in the search for the most likely targets for disease treatment.

Using the method, the scientists, for example, found a family of genes in ordinary baker’s yeast that has a counterpart in humans and may play a role in certain forms of colon cancer.

The computerized system, described in today’s edition of the journal Nature, comes at a time when scientists from private and government-funded labs are competing to finish a detailed map of all 100,000 or so human genes--the complete set of operating instructions for the human animal, known as the genome.

The same companies are already rushing to protect their most useful discoveries by filing patents by the thousands. But without knowing what the genes actually do, biotechnology and pharmaceutical companies won’t be able to secure their patents or pick out the ones most likely to be of benefit to patients and to the bottom line.

The UCLA-spawned start-up company, Protein Pathways, which opened its doors in June with private seed money, has been meeting with several companies looking for ways to pull the gold nuggets from a growing mountain of genetic information.

The method developed by the UCLA researchers takes advantage of the similarities of proteins from one species to another.

“Virtually every disease is a failure of some protein,” says David Eisenberg, a UCLA professor and a founder of the company, along with UCLA’s Todd O. Yeates, now on leave to run Protein Pathways.

The method takes advantage of the similarities of proteins from one species to another.

Proteins are the workhorses of all living things--building cells, repairing damage, and delivering the energy essential to life. And scientists have long known that proteins that do a job in one species, such as yeast, are often similar to proteins that have similar functions in higher animals, including humans.

The discovery of a protein vital to a disease-causing bacteria can lead to development of a new antibiotic that can block it. Understanding what goes wrong with the proteins in cancer cells can provide new treatments.

But the Human Genome Project and other efforts to map out all the genes of humans and other species are producing so much information about genes and the proteins they manufacture that researchers are swamped.

A commentary accompanying the Nature paper--by Andrej Sali at Rockefeller University--describes the approach developed at UCLA and a rival team as a significant contribution to making sense out of all those details.

Last month, Celera Genomics, which is racing to finish a readout of all the human DNA in the human genome, said it has filed provisional patents on 6,500 gene and gene fragments.

But to complete a valid patent application, the company and its corporate partners would have to know what those genes and the proteins they produce actually do.

That’s where Protein Pathways comes in, said Kenneth E. Goodwill, one of the scientists at the start-up. The computer-driven comparisons provide a way of picking out the best of the new discoveries, he said.

UCLA has applied for patents on the methodology that Eisenberg and his colleagues developed in their federally funded laboratory. Protein Pathways has an exclusive license with the university.

Today’s edition of Nature also includes a related article by researchers from Cambridge, England, and Integrated Genomics Inc. in Chicago. These scientists describe a method very similar to one developed at UCLA.

The articles indicate that the UCLA-Protein Pathways submitted their article a few months before the other group, which could prove important in any patent dispute.