Beauty is in the eyes of the beholder, but it might also be tucked away in a handful of genes.
Using genetic information on nearly 4,400 white adults, researchers found that certain genetic mutations were tied to people's beauty ratings from their peers.
Genes were linked to both women's and men's ratings -- but there were differences between the sexes.
In women, certain gene variants tied to beauty were also linked to BMI, a measure of body mass based on weight and height.
Given modern culture, that's not shocking, according to Matthew Keller, a researcher who was not involved in the study.
"It would be surprising if they found there was no correlation," said Keller, of the Institute for Behavioral Genetics at the University of Colorado Boulder. "The effects of genes are not independent of culture."
Of course, there are genes in the human body that go about their business immune to those influences. But, Keller explained, genes associated with a complex concept like "attractiveness" are not operating that way.
"It's not that they are working directly on the trait -- 'My job is to make you attractive,'" he explained.
If this study were done in a different era or a society with different ideas on beauty, Keller noted, the findings might be different, too.
When it came to men's attractiveness ratings, the study found no genetic correlation with BMI. Instead, there was some genetic overlap with blood cholesterol levels.
Why would that be? The researchers, led by Qiongshi Lu, of the University of Wisconsin-Madison, speculated on a reason: Cholesterol is involved in synthesizing testosterone, and more testosterone might mean higher attractiveness ratings for men.
That sounds plausible, according to Keller.
He added, though, that the findings are "early" and have to be viewed with the proverbial grain of salt. Plus, a trait like attractiveness would have to be correlated with many genes, not only a handful.
"What this study found is the very tip of the iceberg," Keller said.
The findings, published April 4 in the journal PLOS Genetics, are based on genetic information from 4,383 people who were part of a larger, long-running health study. All graduated from Wisconsin high schools back in 1957.
In 2004 and 2008, researchers had some of the study participants serve as beauty judges: They looked back at their peers' high school yearbook photos and rated their facial attractiveness on a scale ranging from "not at all attractive" to "extremely attractive." Each photo was judged by a dozen people.
Overall, the researchers found two chromosome regions, containing a "handful of candidate genes" that were related to people's beauty ratings. When they took a deeper look, they found the genetic overlaps between attractiveness and BMI in women, and between attractiveness and cholesterol in men.
The sex of the judges seemed to matter, too: Some genetic variations associated with ratings from women were also tied to hair color; and some of those linked to ratings from men were connected to skin pigmentation.
Julie White, a Ph.D. student in anthropology at Penn State University, cowrote a perspective piece published with the study.
Like Keller, she stressed that any genes associated with attractiveness are not "deterministic." That is, there is no direct line, where having gene variant 'X' means you're attractive or not.
"Attractiveness is hugely complicated," White said.
There are cultural influences: What was deemed attractive in this study of white, Midwestern Americans may not hold true in a different culture, White pointed out. And there are factors other than physical features -- such as personality.
"We read personality in people's faces," White said, "and personality affects attractiveness."
So no one is talking about tweaking any "attractiveness genes" to make pretty designer babies.
Even if someone wanted to try that, White noted, it's unlikely to be possible, given the complicated relationship between genes and attractiveness.
The U.S. National Institutes of Health has a primer on how genes work.