Originally created 12/18/00

Human genome count disputed



How many genes does it take to make a person? Well, maybe not many more than it takes to make a small plant.

That's the rather humbling indication, anyway, from some recent estimates for one of the most basic numbers in biology.

For years, the usual guess has been that there are around 60,000 to 100,000 genes in the human "genome," the collection of DNA that every person carries. That compares pretty impressively to the 25,500 genes in the small flowering plant called Arabidopsis thaliana or thale cress.

And it towers comfortably above the 19,000 genes in the tiny worm C. elegans or the 13,600 genes in the fruitfly Drosophila.

Scientists still disagree wildly on the human gene number. Their formal betting pool records estimates from 27,462 to 153,478 genes, with an average of about 62,000.

Human Genome Sciences Inc. of Rockville, Md., for example, puts the number above 120,000, based on analysis of chemical instructions from genes for making proteins.

But recently, the leader of the international project that has completed a rough draft of the human genome said he expects the gene inventory to fall between 30,000 and 40,000. "That's a surprisingly small number," added Dr. Francis Collins, director of the National Human Genome Research Institute.

And last June, in the journal Nature Genetics, two teams of scientists independently presented studies with estimates in the range of 28,000 to 35,000 genes.

People may find numbers like these disquietingly close to the totals for the Arabidopsis plant or the C. elegans worm. The worm has only 959 cells in its whole body versus trillions in a human body, after all. And neither the plant nor the worm has to deal with such complex tasks as language or figuring out who really won a presidential election.

But scientists say an organism's biological complexity comes from more than the size of its gene inventory.

For one thing, differences in when certain genes are turned on and off can make a big difference in how two organisms develop.

In addition, an organism is actually built and controlled by the proteins its genes tell it to make. While most people are taught that a single gene encodes a single protein, some genes can give rise to more than one protein. That may be the case with more than half the human genes.

What's more, creatures can grow more complex by making more elaborate structures out of the same basic materials. It's the same way that someone can use Lego building blocks to build a simple house or a replica of the Vatican, notes Gerald Rubin of the University of California, Berkeley, and the Howard Hughes Medical Institute.

After all, he said, a single human brain cell isn't that much different from a fly brain cell, but the human brain has an awful lot more of them.

Rubin has had plenty of practice in thinking about the relationship between gene number and complexity. When he and colleagues announced last year that they'd found 13,600 genes in the fruit fly, many scientists thought that number was too low, he said.

After all, that was fewer genes than the C. elegans worm, which seemed a simpler creature.

But in the end, Rubin figures his work "got them out of the box of thinking there was going to be some correlation between complexity and the number of genes."

As for the human gene count, it's good news for scientists if the number turns out to be "smaller than we had feared," says Phil Green of the University of Washington in Seattle.

Green, who estimates about 35,000 human genes, noted that such lower numbers suggest there will be fewer human proteins for scientists to round up into a "complete parts list" for the body.

And in trying to find the genes involved in human disease or characteristics, the smaller number "means less work to do," he said. "We should be able to get to the answer earlier."

On the Net:

The betting pool: http://www.ensembl.org/Genesweep/