When scientists first read out the human genome 15 years ago, there were high hopes that we'd soon understand how traits like height are inherited. It hasn't been easy. A huge effort to find height-related genes so far only explains a fraction of this trait.
Now scientists say they've made some more headway. And the effort is not just useful for understanding how genes determine height, but how they're involved in driving many other human traits.
At first, these problems didn't seem to be so complicated. The 19th-century monk Gregor Mendel discovered that traits in his garden peas, like smoothness and color, could be passed predictably from one generation to the next.
But Joel Hirschhorn, a geneticist at Boston Children's Hospital and the Broad Institute, says it became evident that most stories of inheritance were not so simple. Height turns out to be a prime example.
"People's height didn't behave like Mendel's peas," Hirschhorn says. "It wasn't like they you had two tall people and they'd either have a tall [child] or a short [child]. Often the child was partway between the parents."
Scientists explained this 100 years ago, when they realized that height was influenced by many genes, and each makes a small contribution.
So when the human genome was sequenced, scientists like Hirschhorn thought they could plumb that data to track all the height genes, and finally understand how height — and in fact most other human traits — are shaped by our genes.
That effort started slowly. But now, Hirschhorn says. "for height there are about 700 variants known to affect height, each of them usually with a pretty small effect on height, usually like a millimeter or less."
That massive global effort has involved studying the genes of more than 700,000 volunteer subjects. Even so, the traits they've found only explain about a quarter of the inherited height factors.
And, frustratingly, for most of those variants scientists have no idea what they actually do.
Mostly the variants crop up in mysterious bits of DNA between genes on our chromosomes. That makes it hard to figure out their roles.
So Hirschhorn and his army of colleagues, who reported on the effort Wednesday in the journal Nature, tried a new tack.
Their study focused only on variants that are directly in the genes themselves. By knowing that the genes do, they can understand better how variants might influence height. For example, one is in a gene that influences hormones that regulate growth.
The variants within genes are uncommon, but some have a remarkably large influence on height.
"We found some that, if you carry them, you might actually be an inch taller or an inch shorter, as opposed to just a millimeter difference that we found with the previous variants," Hirschhorn says.
Scientists are still very far from identifying all the genes involved with stature, but these new findings do help them better understand the natural biochemistry that influences height.
So far most of our understanding of height has come from scientists who study children who have abnormal growth patterns, according to Constantine Stratakis, a pediatrician and scientific director of the National Institute of Child Health and Human Development.
"There are rare experiments of nature that have told us these genes are involved in the regulation of growth," he says. In fact, he discovered one of those rare genes, linked to a trait called gigantism.
"It leads to babies that double or triple their length in the first year of life," he says.
These natural experiments have been most useful for treating height disorders, but Stratakis hopes that eventually the genome-search methods will provide leads for future treatments.
The bigger lesson here is figuring out how the biology of a complex trait like height really works.
Rare variants can sometimes make a big difference, "but most of the time it's all about systems that interact that define how an organism behaves, or grows, or has a disease, develops a trait and so on," Stratakis says. "And although it's humbling to see the complexity, at this point it's not unexpected."
Hirschhorn and his colleagues are expanding their already massive study of 700,000 subjects. That approach has drawn skepticism from some scientists, who think it's a waste of effort.
David Goldstein, a professor of genetics at Columbia University, says an expanded effort could ultimately implicate every gene in existence, and that hardly helps scientists narrow down the biological factors that contribute to height.
It's likely scientists will never be able to figure out what these hundreds of common variants do to influence height, Goldstein says. Instead, a much better strategy is what Hirschhorn used in this latest study: looking for rare variants that pack a big punch.
Hirschhorn is undeterred.
"We probably won't get all of the way to explaining 100 percent of the genetic factors, but in some sense that's not really our goal," Hirschhorn says. "Our goal is to use the genetics to do our best at understanding the biology."
To that end, Hirschhorn and his colleagues are not just looking at height; they're digging into traits that make people susceptible to diabetes and obesity.