Giant genome study finds tiny links between genetics and schooling

Huge study finds few genes, effects that barely register.

Everyone wants an easy answer to the big questions about genetics—is there a gay gene? A gene for autism? What about for motherhood or for murder?

In nearly every case, the answer is no; instead, genetic traits are often determined by many small mutations across the genome that interact with the environment and peoples’ experiences. Finding these genetic differences and interpreting their effects is incredibly difficult. The studies that identify them, called genome-wide association studies, entail searching the entire genome of many individuals for areas that consistently correlate with specific traits.

Casting such a wide net necessitates a large sample size, since hundreds of thousands of genetic markers are being tested. Until recently, the most extensive genome-wide association study in the social sciences involved about 10,000 individuals. A new study detailed in this week’s Science examines the genomes of about 100,000 people across fifteen countries in order to identify genetic markers related to a person’s educational accomplishments.

More specifically, the researchers have identified genetic mutations that are associated with two measures of a person’s educational attainment: their total number of years of schooling and their likelihood of finishing college. The paper’s title, “GWAS of 126,559 Individuals Identifies Genetic Variants Associated with Educational Attainment,” sounds pretty promising.

The real numbers, however, are a little less thrilling. The researchers identified three single nucleotide polymorphisms, or SNPs—genetic sequences where one nucleotide has been substituted with another—that predict these two measures of educational attainment. The SNP associated with an individual’s time in school explains a grand total of 0.022 percent of the observed variance in the population. Yes, you read that right: just over two hundredths of a percent, which corresponds to about one month of schooling.

The SNPs related to whether or not a person finished college aren’t much larger in their impact; the largest effect corresponds to a 1.8 percentage point increase in the likelihood of finishing college. Across two million SNPs, the researchers could explain, in total, about two percent of the variation in individuals’ educational attainment.

Since other estimates suggest that about 40 percent of a person’s educational attainment is heritable, these effect sizes seem tiny. But in the world of genome-wide association studies, these numbers aren’t that small. Remember, these are single mutations that are being tested; that may only be one nucleotide out of over 2 billion that is making a significant and measurable difference in how well-educated a person is.

To put these numbers in perspective, the SNPs with the largest effects on height and BMI account for 0.4 percent and 0.3 percent of the variance in these traits, respectively.

Furthermore, it’s particularly difficult to identify SNPs that have a measurable effect on behavioral traits compared to physical traits, thanks to the stronger effects of environment and personality. Many genome-wide association studies in social science actually fail to find any SNPs that have significant replicable effects.

Despite the tiny effect sizes, this research is a definite step toward understanding the genetic architecture of complex behavioral traits, specifically those related to education. Of course, as the researchers are careful to note, there’s no “gene for education.” But by identifying small points of interest throughout the genome, researchers are making progress in understanding the complex role that our genes and our environment play in our educational outcomes.

Science, 2013. DOI: 10.1126/science.1235488 (About DOIs).