Polygenetic Score Linked To Educational Attainment

MedicalResearch.com Interview with:

Dr. Daniel J. Benjamin PhD Associate Professor (Research), USC, 2015-present Associate Professor (with tenure), Cornell, 2013-2015 Assistant Professor, Cornell University, 2007-2013 Research Associate, NBER, 2013-present Faculty Research Fellow, NBER, 2009-2013

Dr-Daniel Benjamin

Dr. Daniel J. Benjamin PhD
Associate Professor (Research), USC, 2015-present
Associate Professor (with tenure), Cornell, 2013-2015
Assistant Professor, Cornell University, 2007-2013
Research Associate, NBER, 2013-present
Faculty Research Fellow, NBER, 2009-2013

MedicalResearch.com: What is the background for this study?

Dr. Benjamin: Educational attainment is primarily determined by environmental factors, but decades of twin and family studies have found that genetic factors also play a role, accounting for at least 20% of variation in educational attainment across individuals. This finding implies that there are genetic variants associated statistically with more educational attainment (people who carry these variants will tend on average to complete more formal education) and genetic variants associated statistically with less educational attainment (people who carry these variants will tend on average to complete less formal education). But none of these genetic variants had been identified until our 2013 paper on educational attainment. That paper, which studied a sample of roughly 100,000 individuals, identified 3 genetic variants associated with educational attainment, each of which has a very small effect. In the current paper, we expanded our sample to roughly 300,000 individuals, with the goal of learning much more about the genetic factors correlated with educational attainment.

MedicalResearch.com: What are the main findings?

Dr. Benjamin: We found a total of 74 genetic variants associated with educational attainment. As in our earlier paper, we found that each of the 74 genetic variants had an extremely small effect. The variant with the strongest association explained only 0.035% of the variation in educational attainment. Put another way, the difference between people with 0 and 2 copies of this genetic variant predicts (on average) about 9 extra weeks of schooling. We found that the results replicated very well in an independent sample of roughly 100,000 individuals from the U.K. Biobank.

Although the effects of the individual genetic variants are small, we can use the pattern of results from across the genome to draw a number of conclusions. For example, we found that the genetic variants disproportionately show up in genes and pathways that are known to be involved in brain development. That suggests that other genetic variants we found – whose functions are not yet well understood – may be good candidates for also being involved in brain development. We also found evidence of genetic overlap between educational attainment on the one hand and Alzheimer’s disease, bipolar disorder, and schizophrenia on the other hand. The genetic variants we identified can therefore be used as candidates for association with psychiatric disorders, so our results may turn out to be useful in understanding the genetic causes of the disorders.

For me, as a social scientist, the most exciting result is that we now know how to construct a “polygenic score” – an index of genetic variants from across the genome – that predicts 3% of the variation across individuals (or 6% if we combine the replication data with the discovery data). That’s not large enough to be useful for predicting any particular individual’s educational attainment, but it’s important because it is large enough to be useful in social science studies, which focus on average behavior in the population. For instance, the polygenic score for educational attainment can be used to study how environments increase or decrease the effects of genetic factors – and we have a preliminary analysis along those lines in the paper.

MedicalResearch.com: What should readers take away from your report?

Dr. Benjamin: Here’s a try for a take-away message: Individual genetic variants have only tiny influences on educational attainment, but studying the many genetic variants that jointly influence educational attainment is nonetheless very useful scientifically.

For example, it can enable more direct studies of how environments increase or decrease the role of genetic factors that influence education attainment; it can lead to insights about the biology of brain development; and it can uncover candidate genes for genetically overlapping traits, such as neuropsychiatric conditions.

MedicalResearch.com: What recommendations do you have for future research as a result of this study?

Dr. Benjamin: There is a lot of future research that I hope will be done. For instance, our polygenic score could be used to conduct careful studies of how genetic and environmental factors interact in influencing educational attainment. As example of how the results from our earlier paper were used, we found that our earlier polygenic score for educational attainment had some predictive power for dementia in older individuals, and several groups of medical researchers have used the genetic variants and polygenic score identified by our earlier study on educational attainment to study other health conditions including dyslexia and psychiatric disorders. By making the results of our analyses publicly available on the our website (the website of the Social Science Genetic Association Consortium, www.thessgac.org), we hope to facilitate similar research that could reach more accurate conclusions by using the more powerful results from our current paper.

MedicalResearch.com: Is there anything else you would like to add?

Dr. Benjamin: Since our work has the potential to be misinterpreted, we wrote a Frequently Asked Questions (FAQ) document to accompany the paper that we’ve posted on our website (http://ssgac.org/documents/FAQ_74_loci_educational_attainment.pdf), which explains not only what we found but also what we did not find. (I’ve borrowed from the FAQ in formulating the responses to some of these interview questions.)

As we wrote in the FAQ, I want to emphasize that we did not find “the genes for” educational attainment—or for anything else. Characterizing the results this way is misleading for many reasons.

  • First, educational attainment is primarily determined by environmental factors, not genes.
  • Second, the explanatory power of each individual genetic variant that we identify is extremely small.
  • Third, environmental factors are likely to increase or decrease the impact of specific genetic variants.
  • Finally, genes do not affect educational attainment directly. Rather, genes that are associated with educational attainment might influence many different biological factors that in turn affect psychological characteristics that finally influence educational attainment.

MedicalResearch.com: Thank you for your contribution to the MedicalResearch.com community.

Citation:

Aysu Okbay et al. Genome-wide association study identifies 74 loci associated with educational attainment. Nature, 2016;

DOI:10.1038/nature17671

Note: Content is Not intended as medical advice. Please consult your health care provider regarding your specific medical condition and questions.

More Medical Research Interviews on MedicalResearch.com